1. Introduction
From time to time, we can feel overwhelmed by our own emotions, and sometimes we only feel relief when we find some control over them. This control can be achieved through the regulation of emotional experience [
1], such that our current emotional system or its expected state becomes able to match our emotional goal. Distraction and reappraisal are two strategies of
emotional regulation (ER). Distraction can be conceived as finding something that takes our mind off the problem situation, whereas reappraisal allows us to see the situation with a fresh pair of eyes. Distraction consists of disengaging attention from the salient aspects of an emotional-eliciting event [
2] by focusing on another concurrent task [
3]. Conversely, reappraisal implies directing attention to the emotionally eliciting event. Since the emotional impact of an event is mediated by the way in which we interpret or appraise it, the aim of reappraisal is to reinterpret and change the meaning of that event [
4].
The efficacy of ER does not exclusively depend on our will. ER through reappraisal needs resources such as time and cognitive capacity [
5]. Similarly, distraction involves multiple focuses on environmental stimuli [
6]. The behavior we use to distract ourselves or reappraise a given situation may, thus, be crucial for the effective implementation of these strategies.
Music listening is one behavior that is recognized as a powerful tool or tactic for regulating emotions [
7,
8,
9]. Music is known to increase well-being and impact the intensity and expression of emotions [
10]. Music listening is used deliberately for several reasons, including the attempt to control, support, or change the current state of experienced emotions [
11]. According to the GSTM (goals, strategies, tactics, and mechanisms) framework [
8], music listening is frequently referred to as a tactic that supports the implementation of distraction and reappraisal [
7,
9], and it seems to be particularly associated with reappraisal. For instance, the message and meaning transmitted by the lyrics of a song positively affect the problem solving and the appraisal of a difficult situation [
12]. Nevertheless, potentially destructive uses of music, like ruminating [
9] and immersing in negative emotions [
12] have also been documented. Thus, music listening seems to be an appropriate tool for ER. However, several questions regarding the relation between music listening and ER are still open.
First, it remains unclear whether music listening is a useful tactic for implementing ER strategies. Associations between music listening and ER have been reported in empirical studies (e.g., [
12,
13]), but most are based on retrospective accounts. Moving away from this retrospective logic, the study of Baltazar et al. [
14] focused on the efficacy of music in stress reduction when combined with an ER strategy, considered relevant by participants, in the context of real-time ER. This study showed that combining music listening with an adequate strategy did not increase ER efficacy, when compared with music or strategy alone. However, the fact that participants were asked to choose the more adequate ER strategy elicited a variety of strategies, including potentially negative ones such as rumination (i.e., thinking repetitively and passively about a particular emotional event [
15,
16]). In addition, the authors did not consider two important individual variables with a potential moderating role in the relation between music listening and ER. The first one is
musical sophistication [
17], conceived as the significance of music in participants’ lives. The second is the level of
executive functioning of participants. Executive functions are a set of abilities that allow decision making, controlling impulsive behaviors, and cognitive flexibility [
18]. The potential moderating role of musical sophistication in music listening-based ER seems straightforward. If music is irrelevant in one’s life, it is unlikely that it will boost ER. To our knowledge, this hypothesis remains untested. Concerning the possible impact of executive functioning (EF) abilities, it has been hypothesized that EF abilities improve ER skills [
19,
20]. Viewing music listening as a tool that aids ER [
7,
9] suggests that music listening may compensate for weak EF skills. Therefore, music listening should be more useful to individuals with lower than for those with higher EF skills. Again, to our knowledge, little is known about this possibility.
Moreover, the chance that music helps the implementation of ER strategies like reappraisal raises questions about the underlying process. One possibility is that music improves empathy. Listening to music has been associated with helping behavior [
21]. Specifically, there seems to be an association between listening to music with prosocial lyrics and an improvement in helping behavior. This suggests that this kind of music can influence our cognition and affect, and it is in line with the media effects theory [
22]: according to this theory, listeners’ thoughts and feelings are congruent with the message conveyed by music. Since prosocial behavior implies some degree of empathy [
23], it can be hypothesized that the key role of music in reappraisal is related to an improvement of empathy. To our knowledge, this hypothesis has not yet been tested.
In the present study, our first goal was to determine whether the effect of music listening on ER is moderated by musical sophistication and/or executive functioning. As a second goal, we wanted to better understand the role played by music in the chosen reappraisal mode—specifically if music promotes empathy-based forms of reappraisal.
To address these goals, we conducted a mixed 2 × 2 × 2 factorial design study with ER type as a within-subjects factor (passive and active), and ER strategy (distraction and reappraisal) as well as music listening (music and no music) as between-subjects. Participants went through a passive ER attempt (PER, no regulation instructions) and an active ER attempt (AER, instructions to regulate). The two ER attempts followed an anger induction task based on autobiographical memories. Before and after anger induction based on an autobiographical memories task, participants self-rated their anger level. In the active condition, participants were randomly assigned to four experimental conditions resulting from the combination of regulation strategy (distraction vs. reappraisal) and music (with vs. without music): distraction with music, distraction without music, reappraisal with music, reappraisal without music. Success in regulation—passive and active—was measured based both on anger ratings and self-reports of perceived success in regulation. To address the first goal, we focused on success in ER during the session (active minus passive regulation) as a function of music listening and regulation strategy to see whether music listening alone influenced ER. Afterwards, we added musical sophistication and executive functioning as moderators, using two new models. In line with the literature and our own analysis of factor structure (see below, methods), executive functioning was approached as a non-unitary construct, where executive functioning tasks do not always show associations among them- mostly in adults [
24,
25,
26]. To address the second goal, we asked participants with vs. without music to report how they reappraised the situation.
Regarding the first goal, we hypothesized that musical sophistication and executive functioning levels would have a moderating effect on ER, meaning that the positive impact of music listening on ER would increase with musical sophistication, and that participants with lower executive functioning skills would be more sensitive to the facilitating effects of music. Concerning the second goal, our hypothesis was that music listeners would implement a form of reappraisal different from the one used by non-listeners. Specifically, music listeners would rely more on empathy.
3. Results
3.1. Control Analyses
Comparisons across the four groups were carried out for emotional regulation abilities and strategies (
Table 2). Results showed no significant differences.
Concerning STAXI-2 values across the experiment, the repeated measures ANOVA showed statistically significant differences across the six time points,
F (5235) = 43.885,
p < 0.001, thus justifying further local comparisons. Regarding anger induction (
Table 3), one sample
t-tests showed that it was significantly above zero in anger induction 1 (
M = 0.699,
SD = 0.649),
t(47) = 7.45,
p < 0.001,
d = 1.08, and anger induction 2, (
M = 0.628,
SD = 0.679),
t (47) = 6.40,
p < 0.001,
d = 0.923. As for the effect of passive vs. active ER tasks, the results from one sample
t tests showed that AER was significantly above zero,
t(47) = 6.80,
p < 0.001,
d = 0.981, meaning that there was a change between pre- and post-active ER moments (
M = 0.580,
SD = 0.591). PER was not significantly different from zero,
t(47) = −0.89,
p = 0.374,
d = −0.130, meaning that there was a non-significant negative change (
M = −0.061,
SD = 0.471) between pre- and post-passive ER moments.
Regarding the presence of Stroop effect, results of a paired sample t-test showed that the number of ink colors in neutral words read by participants (M = 59.3, SD = 16.1) in 45 s was significantly higher than the number of ink colors of negative words (M = 52.2, SD = 15.3), t(47) = 6.64, p < 0.001, d = 0.945, indicating the presence of a significant emotional interference of negative words, and thus a significant Stroop effect. We did not find a significant difference for the positive words (M = 60.8, SD = 15.7) compared to neutral words (M = 59.3; SD = 16.1), t(47) = -1.39, p = 0.171, d = 0.211.
Concerning switch costs and as a result from the paired samples t-test, we found that the global RTs on switch trials (M = 2483, SD = 1061) were significantly longer than the global RTs on repetition trials (M = 2090, SD = 7163), t (935) = −18.4, p < 0.001, d = 0.077, confirming the presence of switch costs.
3.2. Music Effects on ER
Concerning the subjective experience of music effects on ER, Chi-squared tests indicated that, in the distraction condition (n = 24), responses to the question “Did you feel difficulties in the (active) emotional regulation task?” did not differ as a function of listening to music, χ2(1, 24) = 0.098, p = 0.754. Conversely, in the reappraisal condition (n = 24), participants who listened to music showed a significantly larger number of “no difficulties” responses, χ2(1, 24) = 6.171, p = 0.013.
Regarding music effects as measured by STAXI-2 changes, the basic model (ER type*music*strategy) showed a significant main effect of ER type (
Beta = − 0.68;
CI = −1.11–−0.25;
p = 0.002,
Table 4), showing a generalized success in ER (AER more successful than PER). There were no significant interactions, indicating no evidence in favor of music or strategy effects on ER intensity.
3.3. Music Effects on ER Moderated by Musical Sophistication
The model adding Gold-MSI as a fourth factor (ER type*music*strategy*Gold-MSI) showed improved fit compared to the basic model (χ
2(8) = 22.5,
p = 0.004, AIC = 161.02 vs. 167.53). Along with a main effect of ER type (
Beta = 2.37;
CI = 1.13–3.60;
p < 0.001,
Appendix C,
Table A1), we found several significant interactions, including the highest-level interaction ER type x music x strategy x Gold-MSI (
Beta = −0.05;
CI = −0.09–−0.01;
p = 0.009). We broke down the latter by analyzing high- vs. low-Gold-MSI participants, divided according to a median split.
Low-Gold-MSI (
Figure 2) showed a significant interaction for ER type x strategy (
Appendix C,
Table A2). Further analyses considering distraction and reappraisal separately showed significant effects of ER type (significant ER efficacy) in
reappraisal (
Beta = −1.16;
CI = −1.84–−0.48;
p = 0.001), but not in
distraction (
Beta = −0.32;
CI = −0.78–0.15;
p = 0.178). None of the two analyses showed significant interactions between ER type and music. In sum, Low-Gold-MSI participants were able to achieve an effective regulation only by using reappraisal, but music was irrelevant for success.
High-Gold-MSI participants (
Figure 2) showed a significant main effect of ER type. All the interactions were significant (
Appendix C,
Table A3), indicating a relevant role for music. We broke down the highest-level ER type x music x strategy interaction by conducting four analyses, one per music x strategy condition (D-W, D-Wo, R-W, R-Wo). All four analyses showed significant effects of ER type (D-W:
Beta = −0.51;
CI = −0.90–−0.13;
p = 0.008; D-Wo:
Beta = −1.51;
CI = −2.87–−0.16;
p = 0.028; R-W: Beta = −0.68;
CI = −1.12–−0.24;
p = 0.002; R-Wo:
Beta = −0.50;
CI = −0.82–−0.17;
p = 0.002). However, looking at effect sizes as indicated by estimates, we saw different patterns for distraction vs. reappraisal: in
distraction, participants showed an increased effect size (
Beta = −1.51)
without music (vs. with music,
Beta = −0.51), while in
reappraisal the effect size was larger
with music (
Beta = −0.68, vs. without
Beta = −0.50).
In summary, while music was irrelevant to low-Gold-MSI participants, it had some influence on those with higher musical sophistication levels. In the latter, music benefitted reappraisal but harmed distraction.
3.4. Music Effects on ER Moderated by Executive Functioning
The best model among all comparisons engaging up to two EF variables was the one with working memory as measured by digit span (DS) and NA.N (negative non-affective switch costs) as covariates (
Supplementary Materials S3). Low NA.N means lower switch costs, hence high effective flexibility. The analysis showed a significant ER type*music*working memory interaction (
Beta = −0.67,
CI = −1.19–−0.16,
p = 0.011,
Appendix D,
Table A4). Therefore, we broke down the analysis into low- vs. high-working-memory participants (ER type*music*strategy*affective flexibility), based on a median split. Due to the characteristics of the scores (integers, giving rise to several instances of median = 4), two unequal groups were generated (low DS,
n = 44; high DS,
n = 52).
Low DS participants (see
Figure 3) showed a significant interaction between ER type, music, and strategy (
Beta = −1.25,
CI = −2.46–−0.03,
p = 0.044,
Appendix D,
Table A5). We, therefore, broke down the analysis further, considering distraction vs. reappraisal. For
Low DS/distraction, ER type interacted significantly with music (
Beta = 1.11,
IC = 0.20—2.03,
p = 0.017). Breaking down the analysis further, we observed significant ER intensity for both Low DS/distraction with music (
Beta = −0.47,
CI = −0.85–−0.10,
p = 0.013).) and Low DS/distraction without music (
Beta = −1.59,
CI = −2.66–−0.51,
p = 0.004).), but the estimate was larger for distraction without music. In contrast,
Low DS/reappraisal participants showed a significant effect of ER type (
Beta = −0.74,
CI = −1.13–−0.36,
p = < 0.001) without further interactions. Thus, for participants with low DS (low working memory resources), music seems to be relevant when they try to distract, but not when reappraising, and distraction seems to decrease with music.
High DS participants showed a significant ER type*music*affective flexibility interaction (
Beta = −0.00,
IC = −0.00–−0.00,
p = 0.012,
Appendix D,
Table A6). We split the sample into high- vs. low-affective-flexibility individuals, calculating a model for each group.
High DS/high-affective flexibility (
Figure 3) participants showed a significant effect of ER type (
Beta = −0.77,
CI = −1.50–−0.03,
p = 0.04) without further interactions. In contrast, the
High DS/low-affective-flexibility (
Figure 3) showed an interaction between ER type, music, and strategy (
Beta = 0.78,
CI = 0.15–1.42,
p = 0.016). We broke down the analysis further by strategies and found that the
High DS/low-affective-flexibility/distraction group was affected by a significant ER type*music interaction (
Beta = −0.95,
CI = −1.44–−0.45,
p = < 0.001). Splitting the analysis by music, we found significant ER type effects in both conditions (With music:
Beta = −1.10,
CI = −1.68–−0.52,
p = < 0.001; Without music:
Beta = −0.15,
CI = −0.26–−0.04,
p = 0.006), but an increased effect size when music was present. As for
High DS/low-affective-flexibility/reappraisal, the effect of ER type was non-significant (
Beta = −0.18,
CI = −0.53–0.16,
p = 0.304).
In summary, while music seemed to be irrelevant for participants with high working memory and high affective flexibility, it showed a relevant role for those with high working memory and low affective flexibility in distraction, which was carried out more successfully with music.
3.5. Reappraisal Mode with vs. without Music
First, we tried to identify the answers that fit into our predefined category: empathy. Empathy concerns the attempt to understand the other person’s feelings and emotions and comprehend nonjudgmentally the negative experiences of another self [
45]. Some authors refer to two aspects of empathy: cognitive and affective. Cognitive empathy refers to the ability of a person to perceive things from the other’s point of view. Affective empathy concerns the human ability to vicariously share an emotion [
46].
From the remaining answers, four categories emerged (see
Figure 4): positive side/learning, relativization, detachment, and justification. In the positive side/learning category, participants tried to extract the positive aspects of the emotional situation and focus on what they learned from that. When they relativized, participants tried to introduce new elements that mitigate the importance of that situation and look at the situation in a non-absolute way. In detachment, participants tried to distance themselves from the emotional situation, seeing it from an outside point of view. Finally, in justification, participants tried to find a justification for the emotional situation.
The features that participants focused on in each reappraisal mode can be consulted in
Appendix E.
In the R-W condition, participants dominantly reappraised the situation in empathic ways. There seemed to be an increase in cognitive empathy. In contrast, participants in the R-Wo condition focused mainly on trying to see the positive aspects and what they had learned to reappraise the situation.
When participants used empathy to reappraise the situation—dominant reappraisal mode under music—(n = 8), six of them (75%) stated that, in their perspective, the levels of anger would be lower if they remembered the situation again. When they focused on the positive aspects of the things they learned—the dominant reappraisal under no music (n = 6), only two of them (33.3%) said that the feelings of anger would not be the same. Thus, the reappraisal mode associated with music seemed to have been perceived as having more long-lasting effects than that associated with the absence of music.
As for the other question concerning the long-term efficacy of reappraisal (Do you think you will be able to deal better with the emotional situation in the future?), only one participant (who tried to relativize the situation) responded in a negative way. Relativization was not dominant with music nor without music.
4. Discussion
This study had two main objectives. First, we aimed to test the facilitating effect of listening to music on the implementation of two specific ER strategies, distraction, and reappraisal, and if this effect is moderated by musical sophistication and/or executive functioning skills. Second, we aimed to determine whether music listening influences the way in which reappraisal is made, specifically whether music listening promotes empathic reappraisal modes.
4.1. Facilitating Effect: Does Music Make it Easier?
We started by testing the effects of listening to music, without considering musical sophistication or executive functioning as covariables. In line with Baltazar et al. [
14], we observed no evidence that music facilitates the effectiveness of regulation strategies as measured by anger reports (objective measures of regulation). However, subjective measures (perceived success in emotional regulation) showed that participants who tried reappraisal with music reported fewer difficulties than those who did not listen to music.
There are several potential explanations for this apparent contradiction between the two measures of emotional regulation (objective vs. subjective). One relates to problems in objective measures of regulation and includes the possibility that the instrument used to measure state-anger (STAXI-2) failed to capture anger changes properly. For example, the instrument contained an item (“I feel bored”) that attempted to target the way participants felt at that moment due to remembering the annoying event. However, it is not impossible that participants reported their level of boredom resulting from participating in the experiment, instead of referring to the evoked emotional situation. Another explanation relates to possible frailties of subjective reports. The Social Cognitive Theory [
22] postulates that music listening is guided by beliefs about the effects of music on emotional regulation. These beliefs can be created by experience and vicarious learning [
9]. Therefore, if our participants used music to reappraise their emotional situation prior to the experiment, they might have gone through the session believing that music would help in ER, even if it did not. Future research could take into account participants’ beliefs related to music and emotional regulation. It could also consider using another self-report anger measure.
4.2. The Moderating Role of Musical Sophistication
Consistent with the idea that music is more relevant to higher than to lower musical sophistication individuals, our first finding was that music effects on objective measures of regulation were restricted to higher musical sophistication participants. Listening to music made a difference in the ER of higher musical sophistication participants for both distraction and reappraisal, but the effect of music was opposite across strategies: while music benefitted the implementation of reappraisal, it harmed the implementation of distraction. How can we account for this detrimental effect of music during distraction? One possible explanation is that, besides the need to be interesting/relevant, a distractive strategy should be complex and/or unexpected [
6]. It is possible that individuals with higher musical sophistication tend to become increasingly selective and require very high levels of complexity or surprise to achieve distraction. Another possibility is that musically sophisticated individuals are more susceptible to the emotions induced by music [
47] and are, therefore, more susceptible to emotional overloads that make the distraction task harder. According to Patston and Tippett [
48], musicians show decreased performance in cognitive tasks during music listening. Also, in line with this idea of a trade-off between cognitive processing and emotionality, it has been shown that increases in cognitive load during a distracting activity can be related to decreased activation of the emotional system, and vice versa [
6]. Regarding the positive effects of music in the reappraisal condition—which were limited to high-musical sophistication participants, it should be noted that these may not be fully in line with available evidence coming from music-therapy studies. For instance, [
49] found that Cognitive Behavioral Therapy (CBT)—a therapeutic approach where reappraisal has a prominent role—combined with music is effective at relieving anxiety in students. However, the authors did not compare CBT alone with CBT combined with music (we do not know which component was critical: CBT or its combination with music), nor did they measure the musical sophistication of participants (they could be all musically sophisticated, in line with our findings). As for the null effect of music on the emotional regulation of less musically sophisticated participants that we found, it may at odds with evidence that music therapy sessions can be helpful to induce distraction in the elderly during clinical exams or therapy [
50], though, once again, no measures of musical sophistication were presented in this study, and we cannot rule out that participants were highly musical. In general, our findings highlight the importance of considering musical sophistication when it comes to approaching emotional regulation aided by music. Though according to our findings, musical sophistication may modulate the outcomes of emotional regulation interventions, it has not often been considered, at least as much as it has been in studies addressing cognitive improvement (e.g., [
51]).
4.3. The Moderating Role of Executive Functions
Taking the best model containing executive-functioning-related factors, we observed that working memory and a specific domain of affective flexibility (NA.N, the ability to switch into a neutral mode after being exposed to negative stimuli) moderated the effects of music.
Starting with working memory, participants with lower working memory levels showed detrimental effects of music, while those with higher levels showed benefits. In lower working memory participants, music made distraction more difficult. One explanation may be that lower working memory participants have more difficulty in keeping a goal in mind [
52] the goal of achieving distraction. Because music promotes emotions and mind-wandering [
6,
53] music may have driven participants’ attention back to the problem (emotional situation) they were trying to ignore. Conversely, high working memory participants may have been more able to pursue their goal of getting distracted and avoiding problematic memory.
Music was either irrelevant or beneficial for distracting participants with higher working memory abilities. These different effects (null vs. beneficial) depended on effective flexibility. When effective flexibility was high, music was irrelevant; when low, it was beneficial. Unlike the previous moderating effect of working memory, the moderating effect of affective flexibility is in line with our hypothesis. When both working memory and affective flexibility were high, music was irrelevant maybe because participants had all the conditions to distract themselves without needing extra tools like music. In contrast, when participants had greater difficulties in changing into a non affective rule (focusing on the neutral features of the stimulus, in the presence of a negative stimulus), they were able to better distract themselves in the presence of music. The reason why music compensates for the lack of this executive functioning skill is not completely clear. One possible explanation is that the difficulty to focus on the neutral features within the negative stimulus can be overcome by focusing on the neutral features of music. According to this, participants in the distraction with music condition would possibly be focusing on the more neutral aspects of music, like rhythm.
In sum, we found evidence against, but also in favor of our hypothesis. While low levels of working memory do not seem to make participants use music as a support tool, low levels of affective flexibility in a specific domain have the effect we predicted, provided that higher levels of working memory are available. These results should be investigated further with larger and more controlled samples regarding executive functioning.
4.4. Reappraisal Mode: Does Music Make It Different?
Regarding the second objective of this study, we observed differences between music-related conditions in the way participants reappraised the emotional situation. Most participants used empathy-based reinterpretations in the reappraisal with music conditions. It is well-known [
21] that prosocial songs exert an influence on participants’ feelings and thoughts, leading to increased empathy. Interestingly, in our study, we found that the musical features that participants engaged with when using reappraisal were mostly the feelings and memories evoked by the song. Although all songs had lyrics, lyrics were not the dominant driver of empathy (
Appendix E).
The reason why music per se, without words, could generate empathy is not clear. Empathy is a process and not an emotional state, and various components in music may lead to a greater sense of empathy, such as mimicry and cognitive processes [
53]. Mimicry concerns the predisposition to mimic the perceived emotional states of others, increasing, in turn, the predisposition to empathize with them. When listening to music, we tend to mimic the emotional state expressed by the song [
54]. Considering that this mimic-related process is automatic and does not require higher-order cognitive processes [
54], it is possible that mimicked emotions occur. Another potential explanation is that music increases the efficacy of cognitive processes underlying empathy, like perspective taking and inferences about the other’s mental state or intentions. Through visual imagery, music can lead to the creation of scenarios that make the emotional situation more prone to reappraisal. This visual imagery can co-occur with episodic memory, allowing people to visualize a specific event in their lives. These specific events, evoked by music, are usually related to social relations and can serve a nostalgic function [
55]. In such cases, music could lead to the remembering of past positive events with the person that is now involved in the anger situation, leading to a better understanding of the situation and the way things happened the way they did. Future research could investigate these hypotheses, contributing to shed more light on the relationship between music and empathy. It is known that music therapy allows anger management and the improvement of coping strategies in forensic psychiatric patients [
56], which might be related to the link between music listening and empathy that our study highlighted.
Participants who reappraised the situation in an empathic way answered yes more frequently to the questions “In the future, if you remember that emotional situation again, the levels of anger evoked by her would be the same?”. Future research could test further these preliminary results by investigating if this effect does extend in time and the reappraisal of that situation worked upon integration of participants’ life narratives, or, on the contrary, if these feelings of empathy and understanding are only evoked when listening to music. Even if participants who reappraised the situation in a more emphatic way were those that listened to music, this does not mean that there is a correlation.
4.5. Other Limitations and Prospects
A major limitation of our study concerns anger measures. Our objective measures of regulation success were based on self-reports of anger. Self-report measures may be biased by social desirability or demand effects [
57], so future studies should include psychophysiological measures of anger in their experimental procedure.
In addition, we did not control for the possibility that some participants did not have enough memory skills to evoke the feelings of anger, resulting in lower self-reported anger, and lower ER intensity. Even when participants had a personal situation to recall, the fact that they had to recall it twice could have led to a decrease in their emotional impact, resulting in contamination on ER intensity. The two latter limitations can be avoided in the future by using non-personal stimuli to induce emotions, such as emotional pictures or films.
Although we have explicitly asked participants to recall an event that made them feel angry, those events could have elicited other emotions that were not covered by the STAXI-2, leaving some emotional experiences unnoticed. So, future research could include self-reported measures that can evaluate a more diverse spectrum of emotions.
Concerning sample size, we recruited enough participants to capture a medium effect (see methods, a priori power analysis). Though this is a reasonable approach to power, a larger sample would increase the sensitivity of the design to small effects, which may have gone unnoticed in the present study. It is also worth noting that our sample consisted mostly of young female participants. Older people tend to be more efficient in regulating their emotions [
58]. Also, the effectiveness of emotional regulatory strategies is less consistent in women due to sex hormone fluctuations [
59]. Therefore, diversifying sample characteristics also stands as a challenge for the future.
4.6. Overview
The present study contributed to clarifying the influence of music on emotional regulation. Our findings agree with the concept of musical affordance [
60]. According to it, the power of music comes from the dynamic interactions between music and the listener, and making effective use of music in emotion regulation requires perceptual, affective, and sensitive skills. So, maybe ER strategies are like a cognitive way of dealing with emotional events, whereas music can be used as a tactic, provided that there are cognitive resources and/or relations with music. In line with this, our findings showed that music was a useful tactic only for those who are more capable to engage with music in a flexible and effective way [
17]. In the same way, music was a relevant tool for those who have trouble switching to a non affective rule in the presence of a negative stimulus but have the working memory skills to use music to mitigate this.
Moreover, our study highlighted the empathic way participants found to reappraising the emotional situation in the presence of music. So, maybe music does not necessarily reduce the emotional activation caused by anger, but it provides context to find a more meaningful interpretation for the emotional event and promotes the subjective experience of success in regulating emotions. As we observed, music seems to enhance the feeling that, in the future, the problem will have a lower effective resonance, especially when the new meaning assigned makes us understand and forgive others. So, maybe the best use of music does not involve the containment of our emotional experience per se, but the creation of an ideal atmosphere that allows us to integrate this emotional experience into our life narrative.