1 Introduction

People differ in how much they enjoy and appreciate the good things that happen to them. In response to positive events or experiences, people can attempt to maintain, enhance, or prolong their positive emotions by engaging in all sorts of behaviors such as sharing their experiences with others, staying in the moment, reflecting on the meaning of the experience, or counting their blessings. Responding to positive experiences in this way—a process known as savoring (Bryant and Veroff 2007)—is a form of emotional regulation that predicts greater levels of happiness and overall well-being (Hurley and Kwon 2013; Jose et al. 2012; Wilson and Saklofske 2018). Some people, however, do not savor the good things that happen in their lives. In fact, people can engage in responses that actively attempt to decrease their positive feelings following a positive event, a process known as dampening (Quoidbach et al. 2010; Wood et al. 2003). Who is more likely to savor positive experiences in their lives, and who is more likely to respond by dampening their positive emotions? In this research, we studied how people respond to positive events that occur while pursing an activity toward which they are passionate and show that savoring and dampening responses are predicted by specific passion varieties.

2 The Dualistic Model of Passion

The dualistic model of passion defines passion as a strong desire to engage in an activity that one likes, values, pursues on a regular basis, and has incorporated into one’s identity (Vallerand 2015; Vallerand et al. 2003). Importantly, the dualistic model conceptualizes passion as a dualistic construct that comes in two varieties. The first, harmonious passion (HP), emerges when an activity is pursued because of its inherent qualities (e.g., a person simply enjoys the passion) and entails a balanced and flexible engagement in an activity that remains under a person’s control. High levels of harmonious passion lead one to engage in a meaningful activity with a sense of personal volition, which allows it to be well-integrated with the rest of one’s life. The second passion variety, obsessive passion (OP), emerges when an activity is pursued because of certain contingencies that are linked with an activity (e.g., a need for self-esteem), and involves pursuing it at the expense of other life activities. High levels of obsessive passion lead one to engage in an activity with a sense of rigidity and to become consumed by the pursuit of activity-related goals. Therefore, although both HP and OP involve a sense of enjoyment of and love for an activity, they entail different ways of experiencing and engaging in a passionate activity (Vallerand 2015).

A key tenet of the dualistic model is that HP and OP should predict more and less adaptive outcomes, respectively (Vallerand 2015). Research has supported this tenet by showing that HP predicts adaptive outcomes such as life satisfaction, positive affect, and flow. OP, in contrast, does not predict these adaptive outcomes, and can even predict more maladaptive outcomes such as burnout, negative affect, and rumination (Vallerand 2015; for a meta-analysis, see Curran et al. 2015). Consequently, considerable attention has been devoted toward understanding the processes that explain why adaptive outcomes are positively predicted by HP and not OP (e.g., Carpentier et al. 2012). We propose that the way in which passionate people respond to positive experiences could represent an additional pathway that links both passion varieties with differential outcomes.

3 Savoring, Dampening, and Passion

People care deeply about their passionate activities and spend a lot of time and energy pursuing them. The positive events that are experienced while pursuing passionate activities, such as victories, milestones, or accomplishments, are therefore excellent opportunities for people to maintain or amplify their positive feelings by engaging in savoring. But based on our current understanding of passion (Vallerand 2015), responses to positive events in a passionate activity should depend on levels of HP and OP. This hypothesis, however, still remains under-studied.

With an HP, an activity is pursued with a mindful awareness of one’s present experiences (St-Louis et al. 2018) and with a greater capacity to become fully immersed in the activity and experience flow (Philippe et al. 2009; Vallerand et al. 2003). By being mindful and completely immersed in an activity, those with strong levels of HP toward an activity should be able to savor the good things that happen (Bryant et al. 2011; Bryant and Veroff 2007; Ritchie and Bryant 2012). Moreover, an HP entails adopting a positive, realistic outlook on past, present, and future events (Vallerand and St-Louis in press; Verner-Filion et al. 2012). This element of HP should also facilitate savoring because savoring often entails focusing on different temporal periods, including positive experiences that are happening at the moment, happened in the past (reminiscing), or might happen in the future (anticipation; see Bryant 2003; Bryant and Veroff 2007). In contrast, with an OP, an activity often conflicts with other life domains and thus makes it more difficult to be mindful and fully absorbed in the activity (e.g., Philippe et al. 2009; St-Louis et al. 2018; Vallerand et al. 2010). This way of engaging in an activity should make it more difficult to savor positive experiences. Overall, HP toward an activity should enhance one’s ability to savor positive experiences, whereas OP should predict less savoring behavior.

Not only should OP predict to less savoring, but it should also predict the extent to which people dampen their positive emotions in response to positive events. Theoretical arguments can be made for OP being either a negative or positive predictor of dampening. On the one hand, passionate activities come to dominate the self-concept of those with high levels of OP, causing people’s emotional experiences, self-esteem, and even life satisfaction to become contingent on performance in the activity (Lafrenière et al. 2012; Mageau et al. 2011; Verner-Filion et al. 2018). Given the high stakes of activity engagement, highly obsessive people may engage in less dampening in response to positive events in order to allow themselves to feel all the positive emotions that can be experienced during good times. On the other hand, it is possible that OP predicts greater dampening. With the stakes of activity engagement being so high, people with high levels of OP may dampen their positive emotions as part of an ongoing effort to control the emotional highs and lows of activity engagement (e.g., Verner-Filion et al. 2018), to avoid raising expectations about future event outcomes, or because positive feelings are perceived as not being useful for the attainment of important long-term goals (Tamir 2009). Also, given that superstitious thinking is a characteristic of OP (Vallerand et al. 2008, Study 1), highly obsessive people may dampen positive emotions because they hold a common dialectical belief that explains cultural differences in dampening responses: enjoying positive events increases the chances of something bad happening (Miyamoto and Ma 2011). Moreover, research studying how passionate people respond to bad times has found that OP predicts lower levels of self-compassion (Schellenberg et al. 2016), a response pattern that, like dampening, prevents or stifles good feelings (e.g., Neff 2009). Therefore, we reason that there are theoretical arguments that can be made for OP being either a negative or positive predictor of dampening.

One factor that may affect the extent to which both passion varieties predict responses to positive events is the type of positive event that takes place. Recently, an important distinction has been made between positive experiences that occur at two stages of goal pursuit: in-progress achievements and completed achievements (Schall et al. 2017). In-progress achievements are milestones that are reached in the pursuit of an ultimate goal (e.g., a sports fan’s team wins the league semi-final, a student aces a mid-term exam, a novelist writes a chapter). In contrast, completed achievements occur when the ultimate goal is attained (e.g., a sports fan’s team wins the league, a student receives an A+ in the course, a novelist publishes a book). Achieving a milestone in pursuit of a final goal signals goal progress and can be an accomplishment that is worthy of being savored. However, such an in-progress achievement does not guarantee that a final goal will eventually be attained, which may compel people to dampen positive emotions to avoid raising hopes or expectations. Achieving an ultimate goal should be an event that has the most potential to be savored, and may present fewer reasons to dampen positive feelings. Indeed, previous research has shown that people engage in more savoring following successes resulting from completing a task compared to when a task is still in progress (Schall et al. 2017).

4 The Present Research

We conducted three studies to examine the relationship between HP, OP, and responses to positive events. We expected that HP and OP would be positive and negative predictors of savoring, respectively. To examine the relationship between OP and dampening, we initially adopted an exploratory approach (Study 1) and conducted additional research (Study 3) to replicate the observed effects. In Study 1, passionate sports fans reported how much they generally savor positive events that occur while supporting their favorite team, and also reported the extent to which they responded to a recent team victory by savoring and dampening. In Study 2, we focused exclusively on savoring and tested if it mediated the relationship between both passion varieties and outcomes including well-being, academic satisfaction, and intentions to drop out of university in two undergraduate samples. Finally, Study 3 extended the findings of Studies 1 and 2 by testing if savoring and dampening in a sample of passionate soccer fans were affected by the stage of goal pursuit (i.e., in-progress achievement vs. completed achievement). Exploring potential boundary conditions such as goal stage is important for theory development (e.g., Busse et al. 2017), and allowed us test if the relationships observed from Studies 1 and 2 applied following both in-progress and completed achievements. Institutional ethics approval was obtained prior to data collection in each study, and all participants provided informed consent.Footnote 1

5 Study 1

5.1 Method

5.1.1 Participants

Undergraduate sports fans (n = 321) completed an online survey about their fan experiences in exchange for course credit. Additional participants (n = 4) completed the study but were excluded from all analyses because they reported that they did not answer the survey questions honestly. Participants (157 males, 146 females, 18 either did not report a gender or did not identify as either male or female) ranged from 17 to 43 years old (M = 19.36 years, SD = 2.53 years), and most identified having a White/European ethnic background (56.4%). Participants reported their favorite team and answered the survey questions while thinking of their experiences supporting that team. Most participants were fans of teams competing in hockey (48.9%), basketball (18.4%), or soccer (15.3%), and had been supporting their team for an average of 9.38 years (SD = 5.28 years).

5.1.2 Measures

5.1.2.1 Passion

Participants answered items from The Passion Scale (Marsh et al. 2013; Vallerand et al. 2003) to assess levels of HP (6 items; e.g., “Supporting my team is in harmony with the other activities in my life”) and OP (6 items; e.g., “I have difficulties controlling my urge to support my team”) for supporting their favorite team. Each item was answered on a Likert scale ranging from 1 (not agree at all) to 7 (completely agree). Responses from the both subscales were averaged to create overall HP (α = .88) and OP (α = .87) scores.

5.1.2.2 Savoring Beliefs

Participants completed the Savoring Beliefs Inventory (SBI; Bryant 2003; Bryant and Veroff 2007) as an assessment of one’s capacity to savor positive events. The SBI includes 24 items that assess savoring beliefs at three temporal orientations: savoring the moment (e.g., “I know how to make the most of a good time”), anticipating future positive events (e.g., “I feel a joy of anticipation when I think about upcoming good things”), and reminiscing about past positive events (e.g., “I enjoy looking back on happy times from my past”). Participants answered the items while thinking about how they generally feel about positive events that occur while they are supporting their favorite team. Each item was answered on a Likert scale ranging from 1 (not at all true) to 7 (totally agree). All SBI responses were averaged to create a total savoring score (α = .88), and responses from each subscale were averaged to create momentary (α = .74), anticipating (α = .66), and reminiscing (α = .76) savoring scores.

5.1.2.3 Savoring an Important Victory

Sports fans can experience all types of positive events while supporting their favorite team (e.g., goals, grand slams, double plays, etc.), but there is one type of positive event that is particularly relevant for fans: important team victories. Therefore, to complement reports of general savoring behaviors, participants also reported levels of savoring in response to a recent team victory. First, participants briefly described a recent important victory by their favorite team and indicated how much they enjoyed the event by responding to one item: “This victory was the best thing that could happen” (M = 4.65, SD = 1.71). Including this item, which we adapted from the initial section of the Ways of Savoring Checklist (WOSC; Bryant and Veroff 2007), allowed us to test if associations between passion varieties and savoring an important victory remained after statistically controlling for victory enjoyment. They then responded to items from a shortened version of the WOSC (Bryant and Veroff 2007) that was developed by Jose et al. (2012). This 19-item scale asks participants to report how they responded to a specific positive event in their lives. Participants answered the items while thinking of how they responded when the important victory happened. The short-WOSC assesses two factors: savoring (11 items; e.g., “I looked for other people to share it with”) and dampening (8 items; e.g., “I reminded myself that nothing lasts forever”).Footnote 2 All items were answered on a Likert scale ranging from 1 (does not apply at all) to 7 (totally applies). Responses on these two scales were averaged to obtain savoring (α = .91) and dampening (α = .89) scores.

5.2 Results and Brief Discussion

Table 1 reports descriptive statistics and correlations between HP, OP, savoring beliefs, and responses to the recent team victory.Footnote 3 Of primary interest in Table 1 are the partial correlations between each passion variety (controlling for the other passion variety) and savoring and dampening responses, which are reported above the diagonal.Footnote 4 HP (controlling for OP) positively predicted savoring beliefs at each temporal perspective and savoring responses to a recent team victory. OP (controlling for HP) predicted lower levels of savoring beliefs at each temporal perspective and greater levels of dampening reactions to the team victory. We also found that levels of enjoyment of the team victory were positively correlated with both HP (r = .399, p < .001), and OP (r = .332, p < .001), and that the strength of these correlations did not differ, Hotelling’s t (297) = 1.570, p = .118.Footnote 5 This means that both passion varieties predict different reactions to a team victory, even though they predict similar levels of enjoyment of the victory. Importantly, the partial correlations between passion varieties and both savoring and dampening reactions to the team victory did not change in any meaningful way when victory enjoyment was included as an additional covariate.

Table 1 Study 1: passion varieties, savoring beliefs, and responses to a recent team victory
Full size table

These results provide initial support for the hypotheses that savoring responses are positively and negatively predicted by HP and OP, respectively. HP predicted greater levels of savoring, both in general (i.e., greater savoring beliefs) and in response to a specific positive event (i.e., greater savoring in response to a recent team victory). This means that, with HP, sports fans actively savor positive events and try to make the most out of them. OP, however, not only predicted lower levels of savoring beliefs, but also higher levels of dampening in response to a recent victory. We reasoned that there are theoretical arguments that can be made for OP being either a positive or negative predictor of dampening behavior. The results of this study provide initial support for OP being a positive predictor of dampening, and suggest that sports fans with high levels of OP actively attempt to stifle their positive emotions when good things happen.

6 Study 2

Our aim in Study 2 was to build on the results of Study 1 by focusing exclusively on savoring and assessing its role in the relationship between both passion varieties and outcomes. The associations observed in Study 1 suggest that savoring may play a part in explaining why HP tends to predict adaptive activity-related outcomes and OP does not. Moreover, according to the dualistic model of passion, the positive experiences that accumulate while pursuing a passionate activity allow the effects of passion to extend beyond the passionate domain and affect how people feel in their lives in general (Vallerand 2016). Therefore, savoring within the context of a passionate activity may also explain associations with global well-being outcomes. To test the mediating role of savoring in the relationship between HP, OP, and outcomes both within and beyond the scope of a passionate activity, we recruited two samples of undergraduates who responded to questions about either their passion for their favorite activity or their academic passion. Savoring is posited to contribute to one’s satisfaction in an activity (e.g., Bryant 2003). We thus focused on how passion varieties and savoring related to levels of satisfaction, and assessed indices of satisfaction at two levels: domain-specific satisfaction (i.e., academic satisfaction, intentions to drop out) and global satisfaction (i.e., satisfaction with life).

6.1 Method

6.1.1 Participants

Undergraduates (n = 552) completed an online survey about their favorite activities in exchange for course credit. Additional participants (n = 9) completed the study but were excluded from all analyses because they reported that they did not answer the survey questions honestly. Participants (403 females, 139 males, 10 did not report a gender) ranged from 16 to 47 years old (M = 19.92 years, SD = 3.50 years), and most often identified having a White/European ethnic background (45.5%). Participants were randomly assigned to complete one of two versions of the survey. Participants who completed the general version (n = 273) reported their favorite activity and responded to items assessing passion and savoring in relation to that activity. Participants who completed the academic version (n = 279) responded to items assessing passion and savoring in relation to their academic experiences. Having two versions of the survey allowed us to test if passion varieties and savoring predicted general outcomes related to a self-reported favorite activity (general version), and domain-specific outcomes related to a specific activity (academic version).

6.1.2 Measures

6.1.2.1 Passion

Participants answered items from The Passion Scale while thinking of either their favorite activity (general version) or academics (academic version).

6.1.2.2 Savoring Beliefs

Participants completed the SBI while thinking of their capacity to savor positive events that occur in their favorite activity (general version) or academics (academic version). There were very strong correlations between the three SBI subscales in both the general (rs = .625 to .651) and academic (rs =.647 to .754) versions (all ps < .001). Due to these strong correlations and the strong correlations obtained in Study 1, we focused exclusively on the total SBI score.

6.1.2.3 Outcomes

All participants completed the Satisfaction with Life Scale (Diener et al. 1985) as a measure of subjective well-being. This 5-item scale asks participants to rank their agreement with statements about their overall satisfaction with life (e.g., “In most ways my life is close to my ideal”) on a scale from 1 (strongly disagree) to 5 (strongly agree). Participants who completed the academic version also completed two measures of academic-specific outcomes. First, as a measure of academic satisfaction, participants completed the 8-item school satisfaction subscale of the Multidimensional Students’ Life Satisfaction Scale (Huebner 2001). Participants answered the items while thinking of their satisfaction in university (e.g., “I like being in university”) on a scale from 1 (strongly disagree) to 6 (strongly agree). Second, intentions to drop out of university were assessed by asking participants to report how often they had felt like dropping out of university since the start of the semester. Participants answered three items (“I considered dropping out of university”; “I had some intentions to drop out of university”; “I thought that I would be happier by quitting the university”) on a scale from 1 (never) to 7 (always).

6.1.3 Data Analysis

We used Mplus to test the hypothesized path models. When analysing responses from the general survey, we tested a model in which both HP and OP (independent variables) predicted savoring (mediator), which in turn predicted satisfaction with life (outcome variable). We analyzed the same model with responses from the academic survey, but with academic satisfaction and dropout intentions included as two additional outcome variables. In order to adjust for measurement error, all observed variables were modeled as single indicators of latent variables with loadings fixed at 1.0 and unique variances fixed to values computed using the following formula: (1-reliability) * variance (see Williams and O’Boyle 2008).Footnote 6 All models were tested using robust maximum likelihood estimation (MLR) to account for potential deviations in normality. Evidence of mediation was tested using the 95% confidence intervals of the indirect effect estimated with bias-corrected bootstrapping (5000 samples).Footnote 7

6.2 Results and Brief Discussion

Table 2 reports descriptive statistics and correlations between passion varieties, savoring, and outcome variables from both the general and academic online surveys. Models with the general and academic surveys are displayed in Figs. 1 and 2, respectively. Tests of direct and indirect effects are reported in Table 3.

Table 2 Study 2: passion varieties, savoring beliefs, and outcome variables
Full size table
Fig. 1
figure 1

Study 2: mediation model of relationships between passion varieties, savoring, and satisfaction with life (general version, N = 273). Standardized estimates are reported. *p ≤ .05; **p ≤ .01

Full size image
Fig. 2
figure 2

Study 2: mediation model of relationships between passion varieties, savoring, and satisfaction with life, academic satisfaction, and drop out intentions (academic version, N = 279). Standardized estimates are reported. *p ≤ .05; **p ≤ .01

Full size image
Table 3 Study 2: standardized estimates of total, direct, and indirect effects
Full size table

In the general model, savoring was positively predicted by HP (β = .453, p < .001) and negatively predicted by OP (β = − .346, p < .001). Savoring, in turn, positively predicted satisfaction with life (β = .371, p < .001). Tests of indirect effects revealed that both HP and OP were indirectly related to satisfaction with life via savoring.

In the academic model, savoring was positively predicted by HP (β = .710, p < .001) and negatively predicted by OP (β = − .735, p < .001). Savoring, in turn, positively predicted satisfaction with life (β = .334, p < .001) and academic satisfaction (β = .225, p = .010), and negatively predicted dropout intentions (β = − .203, p = .011). Tests of indirect effects revealed that both HP and OP were indirectly related to each outcome via savoring.

Overall, these results (a) replicated the effects observed in Study 1, and (b) provided preliminary evidence that savoring mediates relationships between passion varieties and both global (i.e., satisfaction with life) and domain-specific (i.e., academic satisfaction, drop out intentions) outcomes. These mediating effects suggest that savoring may be an additional process that explains why HP and OP predict diverse outcomes. We elaborate more on this point in our General Discussion.

7 Study 3

The aims of Study 3 were to (a) replicate the effects obtained in Studies 1 and 2, and (b) delve further into the issue of savoring and dampening by testing if their relationships with passion varieties depended on whether a positive event was a result of an achievement on the journey toward a final goal (in-progress achievements), or the final attainment of the goal itself (completed achievements). In designing Study 3, we relied on hypothetical scenarios to assess how passionate people would respond to either in-progress or completed achievements. Hypothetical scenarios are useful in studying how people expect to behave in different situations (Schall et al. 2017), and have been used to assess savoring in response to an event that is identical for all participants (e.g., Gentzler et al. 2016). Using scenarios also allowed us to experimentally manipulate the status of goal pursuit. To make the scenarios as realistic as possible, we recruited soccer fans who supported specific teams that were, at the time of the study, competing in the UEFA Champions League and asked them to imagine and report how they would respond if their team won the semi-final (in-progress achievement) or the final game of the league (completed achievement). Both scenarios depicted credible outcomes; at the time of the study, each team was either at the top or tied at the top of their respective Champions League group, meaning that they all had a chance of continuing to the semi-final and final stages of the Champions League.

7.1 Method

7.1.1 Participants

Soccer fans (n = 394) were recruited from Prolific Academic (www.prolific.ac), an online platform oriented toward academic research (Peer et al. 2017). Users complete a screening questionnaire upon initial registration with Prolific Academic, and to be invited to participate in the current study they must have reported on the screening questionnaire that they were (a) 18 years of age or older, and (b) fans of one of the following soccer teams: Chelsea (n = 67), Liverpool (n = 84), Manchester United (n = 174), Manchester City (n = 39), or Tottenham Hotspur (n = 30). Additional participants (n = 21) qualified for and completed the present study but reported on an integrity check item that they were not fans of any of these teams; these participants, along with 7 additional participants who reported that they did not answer the survey questions honestly, were excluded from all analyses. Participants (193 males, 188 females, 13 either did not report a gender or did not identify as either male or female) ranged from 18 to 67 years old (M = 30.09 years, SD = 9.70 years), and most identified having a White/European ethnic background (79.9%). Participants had been supporting their team for an average of 12.93 years (SD = 11.17 years). Participants received £0.77 (approximately $1.08 USD) in exchange for participation.

7.1.2 Procedure and Measures

First, participants completed the Passion Scale to assess levels of HP (α = .89) and OP (α = .91) for supporting their team. They were then randomly assigned to read one of two scenarios about their favorite team’s progress in the Champions League. Those randomly assigned to the in-progress condition imagined how they would feel and react in response to the following scenario:

Your favorite team has just won the semi-final of the UEFA Champions League. Your team has played very well over the course of all the Champions League games and has defeated some of the top teams in Europe. You are very happy with your team’s performance. Your team will be competing in the UEFA Champions League final in a few weeks.

Those randomly assigned to the completed condition read the following scenario:

Your favorite team has just won the UEFA Champions League. Your team had played very well over the course of all the Champions League games and had defeated some of the top teams in Europe. You are very happy with your team’s performance.

After reading the scenario, all participants reported the extent to which they would engage in savoring and dampening behavior in response to the scenario. Savoring was assessed with five items (see Schall et al. 2017): “I would savor the present moment”, “I would think about how happy I am”, “I would think about things that made me feel pleasure”, “I would focus all my attention on my happiness”, “I would think about things that help me prolong my positive feelings”; α = .91). Dampening was assessed with three items: “I would think about things that decrease my positive feelings” (Schall et al. 2017), “I would think about things to decrease my good feelings—to make myself feel not as good”, and “I would think about things to decrease my excitement—to calm down” (see Wood et al. 2003; α = .86). All items were assessed on a scale from 1 (not agree at all) to 7 (completely agree).

7.1.3 Data Analysis

Table 4 reports descriptive statistics and correlations between all variables. First, we tested the associations between each passion variety and both savoring and dampening responses by calculating partial correlations; these results are reported above the diagonal in Table 4. Second, we tested if these associations were affected by whether participants imagined their favorite team winning the semi-final or final of the Champions League using moderated regression analyses with the PROCESS macro in SPSS (Hayes 2013). Each regression analysis included (a) the focal passion variety (mean centered), (b) the scenario condition (dummy coded: 0 = in-progress condition; 1 = completed condition), (c) the passion × condition interaction, and (d) the other passion variety (mean centered) as a covariate (PROCESS Model 1). A significant passion × condition interaction would indicate that the association between a specific passion variety and outcome variable was moderated by the scenario manipulation. Unstandardized beta coefficients are reported for these analyses.

Table 4 Study 3: passion varieties, savoring, and dampening behavior
Full size table

7.2 Results and Brief Discussion

7.2.1 Savoring

First, consistent with the results from Studies 1 and 2, HP (controlling for OP) positively predicted savoring, r = .484, p < .001. OP (controlling for HP) also positively predicted savoring, r = .136, p = .007; this relationship, however, was not as strong as the relationship between HP and savoring, Hotelling’s t (391) = 4.522, p < .001. Second, we did not find any evidence that the scenario manipulation moderated the relationship between savoring and either HP (b = − 0.002, SE = .079, p = .985, 95% CI [− 0.157, 0.154]) or OP (b = 0.043, SE = .073, p = .559, 95% CI [− 0.101, 0.187]). These results mean that both HP and OP positively predicted greater levels of savoring in response to both scenarios, although the relationship with savoring was stronger with HP compared to OP.

7.2.2 Dampening

First, HP (controlling for OP) did not predict dampening, r = − .036, p = .474, whereas OP (controlling for HP) positively predicted dampening, r = .477, p < .001. These results replicate those from Study 1 by finding that OP, but not HP, predicts a greater tendency to dampen positive feelings. Second, we found that the scenario manipulation moderated the relationship between dampening and both passion varieties (see Fig. 3). HP predicted decreases in dampening in the completed condition relative to the in-progress condition, b = − 0.204, SE = .094, p = .031, 95% CI [− 0.390, − 0.019]. The simple slopes of HP at each condition, however, did not significantly differ from zero (in-progress condition: b = 0.069, SE = .078, p = .380, 95% CI [− 0.085, 0.223]; completed condition: b = − 0.136, SE = .076, p = .076, 95% CI [− 0.285, 0.014]). OP predicted increases in dampening in the in-progress condition, but this effect was attenuated in the completed condition, b = − 0.256, SE = .087, p = .003, 95% CI [− 0.427, -0.086]. The simple slopes of OP at each condition revealed a significant positive effect at both conditions (in-progress condition: b = 0.733, SE = .070, p < .001, 95% CI [0.594, 0.871]; completed condition: b = 0.477, SE = .072, p < .001, 95% CI [0.336, 0.618]). These results mean that the relationship between both passion varieties and dampening is sensitive to whether the positive event was the attainment of an important goal (completed condition) or a milestone on the journey toward the goal (in-progress condition). High levels of either HP or OP predicted less dampening behavior in response to the scenario described in the completed condition compared to the in-progress condition. However, the main difference in these effects is that, unlike HP, OP positively predicted dampening in both conditions.

Fig. 3
figure 3

Study 3: relationship between dampening and harmonious passion (panel A) and obsessive passion (panel B) for participants randomly assigned to the in-progress and completed conditions. Simple slopes are plotted at 1 standard deviation (SD) above and below the mean

Full size image

8 General Discussion

Good things happen when people pursue their favorite activities: they win games, ace tests, and achieve all types of goals, milestones, and accomplishments. In this research, we found that the extent to which these positive events are savored or dampened depends on the extent to which one’s passion for an activity is harmonious or obsessive. HP predicted a greater capacity to savor (Studies 1 and 2) and greater savoring after a recalled (Study 1) and anticipated (Study 3) positive event, whereas OP predicted less savoring capacity (Studies 1 and 2) and even a greater tendency to dampen positive emotions (Studies 1 and 3). These results reveal that passion varieties matter for predicting how people manage their good feelings following positive events.

Savoring allows people to sustain, prolong, or up-regulate the positive emotions that occur following positive events or experiences (Bryant and Veroff 2007). There are many features of HP that should make it most conducive to savoring: HP entails a mindful awareness of one’s ongoing experiences, a greater capacity to experience flow and concentrate during an activity, and involves engaging in an activity with a secure sense of self-esteem. Conversely, OP should be less conducive to savoring: OP entails lower levels of mindfulness, greater difficulties concentrating and being fully present during an activity, and involves engaging in an activity with a contingent, fragile sense of self-esteem (Vallerand 2015). This research supports the hypothesis that savoring is positively predicted by HP and negatively predicted by OP. These findings have also been replicated recently in a sample of part-time workers (Vallerand and St-Louis, in press). We should note that, contrary to our hypothesis, OP was positively associated with savoring in Study 3. However, this association was relatively small (partial r = .136) and was significantly weaker than the relationship between HP and savoring. Study 3 also assessed anticipated savoring in response to a hypothetical scenario, which could mean that OP predicts anticipated but not actual savoring responses. Overall, it appears that people with high levels of HP try to make the most of good times by engaging in savoring, whereas people with high levels of OP do not savor positive events to the same extent.

People can also respond to positive events by trying to stifle or down-regulate their positive emotions by engaging in dampening. Initially, we proposed that there was a theoretical basis to predict that OP would be associated with either more or less dampening. The results of this initial investigation were consistent and found that OP positively predicted dampening in response to a recalled positive event (Study 1) and a potential future positive event (Study 3). People with high OP may dampen their positive emotions in an effort to maintain emotional stability during activity engagement (Verner-Filion et al. 2018), to facilitate goal attainment (Tamir 2009), or because their superstitious tendencies (Vallerand et al. 2008) lead them to believe that experiencing positive feelings would increase the chances of something bad occurring (Miyamoto and Ma 2011). Another possibility is that many positive events really are more bittersweet and elicit more mixed feelings for those with higher levels of OP. OP involves an all-encompassing, rigid relationship with an activity (Vallerand 2015). This means that OP could predict more dampening because many positive events also signal the end of an activity. For example, for an obsessive sports fan, supporting a team that wins a championship is a very positive event, but this also means that the season is over and that no games will be played for an extended period of time. This bittersweet explanation could clarify why, in Study 3, OP positively predicted dampening responses even when passionate soccer fans imagined their favorite team winning the prestigious UEFA Champions League. Even though the relationship between OP and dampening was not as strong among participants who imagined a finals victory (completed condition) compared to those who imagined a semi-final victory (in-progress condition), it is nonetheless fascinating that highly obsessive sports fans reported that they would respond to this championship victory by trying to decrease their positive feelings.

This research also contributes to our understanding of the self-regulatory pathways that link both passion varieties to outcomes. Past research has focused on the role of coping with adversity in explaining why HP, in general, predicts adaptive outcomes and OP does not (e.g., Schellenberg and Bailis 2016; Schellenberg et al. 2013). But just like people must manage and regulate the negative emotions that accompany stress and adversity (Lazarus 1999), people must also regulate their positive emotional experiences that accompany positive events (Gross 1999). The way people manage their positive emotions, either by savoring or dampening, likely represents another mechanism that connects passion varieties with outcomes. Evidence for this process with savoring was obtained in Study 2, in which savoring mediated relationships between both passion varieties and satisfaction with life, academic satisfaction, and drop out intentions. This means that, like coping during bad times, savoring during good times is an important self-regulatory process that affects important outcomes including satisfaction with a passionate activity and with life in general.

8.1 Limitations and Future Directions

This series of studies should be considered part of an initial effort toward understanding how passionate people respond to positive events (see also Vallerand and St-Louis, in press). There are many ways in which future research can build on these findings. In each study, the relationship between passion varieties, savoring, and dampening was tested using correlational methods, meaning that causality cannot be inferred. Therefore, methods that have been developed to experimentally manipulate HP and OP (see Vallerand 2015) would be particularly useful in this area. Longitudinal designs would also allow for more stringent tests of the mediation effects reported here, and could also test the indirect role of savoring in the relationships between passion varieties and other outcome variables (e.g., performance, health; Curran et al. 2015). We also proposed several reasons for why savoring and dampening are differentially related to both passion varieties (e.g., differences in mindfulness, superstitious beliefs, the extent to which an event is bittersweet). Research is needed to study these proposed mechanisms and test why HP and OP predict savoring and dampening.

Finally, we should note that this research highlights several of the complexities involved in positive emotion regulation. For example, in Study 1 the negative association between OP and savoring at the global level (i.e., savoring beliefs) was not replicated when savoring was assessed in response to a specific positive event (i.e., a recent team victory). This suggests that situational characteristics of positive events (see Bryant and Veroff 2007), such as whether or not positive events are experienced in the presence of others (e.g., Greenaway and Kalokerinos 2017), may affect the relationship between passion varieties and savoring and dampening responses. Also, the results showed that savoring a positive experience did not necessarily mean that dampening was not taking place; in fact, in both Studies 1 and 3, savoring and dampening were positively correlated. This means that people can engage in savoring and dampening at the same time. We can see how this process could unfold with a passionate sports fan who experiences a team championship: the fan could engage in savoring by sharing the win with others and reflecting on the meaning of the victory, while at the same time engaging in dampening by thinking about how this good time will eventually end and about how it is unlikely that the team will win another championship. In our view, these results underscore the need for continued research in this area to better understand the complex relationships between savoring, dampening, and passion.

9 Conclusion

People can respond to positive events in different ways. In this research, we focused on how passionate people responded to the positive events that happened while pursuing a favorite activity and found that levels of HP toward an activity predicted more savoring (i.e., attempting to sustain, prolong, or enhance positive emotions), whereas levels of OP toward an activity predicted less savoring and more dampening (i.e., attempting to stifle positive emotions). These findings show that the ways in which people respond during good times depends on the quality of their passion for an activity. This research also identifies savoring and dampening as self-regulatory processes that could explain why HP and OP predict different domain-specific and global outcomes. Moving forward, research is needed to continue exploring longitudinal and causal associations between passion varieties and how people respond when good things happen.