Unique personality profiles predict when and why sad music is enjoyed

Abstract

The experience of sadness is largely unpleasant, but when expressed through music, it can be pleasurable. Previous research has shown that an attraction to sad music is correlated with personality traits like empathy, Absorption, and rumination. However, the intricacies of the relationship between personality, situational factors, and reasons for engaging with sad music have yet to be fully explored. To address this, participants (N = 431) reported the situations in which they would listen to sad music and their motivations for doing so. Regularized regression models were employed to assess correlations between personality, situational, and motivational factors. Mediation models were used to determine if emotional responses mediated these associations. People who scored higher on Absorption, the Fantasy component of empathy, and rumination reported enjoying sad music. Absorption and Fantasy were associated with liking sad music because of its ability to regulate/enhance positive emotions. Rumination was associated with liking sad music in tense situations because it both strengthens positive and releases negative emotions. Our results further our understanding of reward responses to negative stimuli by highlighting the role of personality and situational factors. Such findings have implications for the development of interventions for mood disorders, in which music could be used as a tool to regulate emotions and re-engage the reward system.

Keywords

sad music emotion enjoyment motivations lasso regression mediation

From an evolutionary standpoint, sadness is considered one of the basic, utilitarian emotions that we seek to minimize. The negative sensations associated with sadness are generally elicited by a real or perceived loss and the resulting bodily-state changed are designed to protect social status during these potentially vulnerable situations (Damasio & Carvalho, 2013). Sadness is also associated with cognitive changes that may serve adaptive functions, such as increasing realistic inference-making (Moore & Fresco, 2012), encouraging perseverance, and enhancing interpersonal relationships (Forgas, 2013). These benefits may explain our implicit understanding that there is inherent value in feeling sad. However, knowing this function of sadness does not fully explain situations that commonly occur in response to music and other forms of art, in which music that conveys and elicits sadness is found to be purely enjoyable. Often referred to as the “tragedy paradox” in the philosophy literature (Oliver, 1993), a deep psychological understanding of the ways in which listening to sad music can become linked with cognitive reward is lacking. Gaining a better understanding of this problem may provide new insights into the function of negative emotional states. It may also resolve inconsistencies regarding the role that such emotions play in the rewarding aspects of musical stimuli and help account for our motivations for engaging with such stimuli.

Individual differences and the enjoyment of sad music

Previous explorations of this tragedy paradox have hypothesized that sadness, when conveyed through music—a purely aesthetic, non-threatening context—can be found pleasurable because it can yield psychological benefits related to mood regulation, recollection of and reflection on past events, and empathic understanding (Levinson, 1997; Sachs, Damasio, & Habibi, 2015; Taruffi & Koelsch, 2014). Several of these psychological rewards appear to occur specifically in response to music that conveys sadness and not in response to music that conveys happiness, for example (Taruffi & Koelsch, 2014). The nature of the cognitive reward may depend on one’s emotional response to expressions of sadness conveyed through the music (Kivy, 1991; Robinson, 1994). Much of the previous literature has focused on personality measures that predict the enjoyment of sad music; however, few studies have directly tested how specific groups of individuals perceive and experience sad music differently and, by extension, how such groups can gain different psychological rewards from the music. Openness to Experience, one of the domains of the Big Five model that encompasses personality dimensions such as aesthetic appreciation and novelty-seeking, has been shown to be positively associated with liking of sad music McCrae (2007): listeners who liked sad music that made them feel sad tended to score higher on Openness to Experience (Ladinig & Schellenberg, 2012; Vuoskoski, Thompson, McIlwain, & Eerola, 2011). Different subcomponents of empathy, such as Fantasy, which refers to the tendency to become transported into the feelings of the characters when engaging with works of fiction, and Empathic Concern (EC), which refers to feelings of sympathy and concern for unfortunate others, have both been found to be positively associated with liking sad music (Vuoskoski et al., 2011). In other explorations, an additional subcomponent of empathy, Perspective Taking (PT), which refers to the tendency to adopt the viewpoint of another, was found to be related to liking sad music (Taruffi & Koelsch, 2014) and that this relationship was mediated by a heightening of the emotional responses to sad music (Kawakami & Katahira, 2015). A personality measure related to Fantasy, called Absorption, which refers to the capacity to become deeply focused on a stimulus and temporarily disengaged from the external surroundings, has also been shown to be linked to the enjoyment of sad music (Garrido & Schubert, 2011)

On the contrary, personality traits and symptoms related to depression, such as a tendency to ruminate, appear to be associated with an attraction to sad music as well. Numerous studies have suggested a link between measures of trait rumination and liking sad music (Garrido & Schubert, 2015; Schubert, Halpern, Kreutz, & Garrido, 2018). Severity of depressive symptoms more generally has also been shown to be positively associated with liking sad music, a relationship that was found to be more pronounced in men than women (Hogue, Crimmins, & Kahn, 2016). However, it is unclear if these traits are linked to feelings of pure enjoyment in response to the sad music, or with liking because of some other cognitive processes that becomes engaged (Garrido & Schubert, 2013). Such findings therefore highlight the distinction between “liking” and “enjoyment,” proposing that personality traits may play a role in which cognitive mechanisms are triggered when listening to sad music.

The role of emotional responses in the reactions to sad music

Despite the support for an association between personality and the liking of sad music, several studies have failed to replicate certain associations. Taruffi and Koelsch (2014), for instance, did not find a direct link between EC and liking sad music; rather, they found that EC was related to feeling sad in response to the music. This suggest that the relationship between personality and liking sad music may be mediated by emotional responses to sad music. For example, given that Openness to Experience and Fantasy/Absorption are correlated with the intensity of emotions felt in response to sad music (Vuoskoski et al., 2011), it has been proposed that listeners who demonstrate these traits can appreciate the emotions associated with music listening in a vicarious way, dissociating from the feelings that typically accompany a negative experience (Garrido & Schubert, 2015).

On the contrary, there is evidence that rumination and depressive symptoms may actually prolong negative emotional states in response to sad music and maintain negative moods (Garrido & Schubert, 2013; Wilhelm, Gillis, Schubert, & Whittle, 2013). A related trait, nostalgia proneness, which refers to the tendency to experience a mix of positive and negative feelings when reflecting on the past, has also been linked to liking sad music because of this increase in negative emotions (Barrett et al., 2010; Garrido & Schubert, 2013). It has been suggested that the link between these traits and liking sad music is representative of a “maladaptive attraction” to negative emotions (Garrido & Schubert, 2015) given that there is some evidence to suggest that the experience is not enjoyable (Wilhelm et al., 2013). Why people would choose to listen to sad music if it does not improve their mood remains unclear.

Collectively, these results suggest that quality and intensity of the emotional responses to sad music appear to play a role in its enjoyment, by either increasing positive emotions or by prolonging negative ones. It is quite possible, then, that the psychological benefits of listening to sad music, and therefore people’s reasons for selecting sad music, depend on the nature of the emotional response. While there is some empirical evidence for this idea, whether emotional responses to sad music mediate role relationship between liking sad music and personality traits has not yet been systematically evaluated.

Situational factors influencing engagement with sad music

Previous research has also suggested that the enjoyment of sad music is influenced by situational factors. It has been shown that when people are listening to music alone, they demonstrate increased emotional responses, as measured by psychophysiology (Egermann et al., 2011), and, relatedly, report choosing to listen to sad music more often when alone (Juslin, Liljeström, Västfjäll, Barradas, & Silva, 2008). Social feedback can also influence the way in which one reports experiencing emotions evoked by music (Egermann, Kopiez, & Altenmüller, 2013). Listening to sad music in a group setting, for example, can lead to heightened ruminative thoughts and increases in negative moods, particularly in participants who are prone to depression (Garrido, Eerola, & McFerran, 2017). Others have reported that people choose to listen to sad music more often when experiencing distress, such as after a death or a breakup (Taruffi & Koelsch, 2014). It has been proposed that situations influence enjoyment of sad music by manipulating one’s mood, though not everyone chooses to listen to sad-sounding music when they are sad. Personality may additionally play a role in whether people choose to listen to music that is congruent with their current mood. People who scored higher on the Personal Distress (PD) subscale of empathy, and lower on the personality trait Emotional Stability, were more likely to listen to sad music when they were in a negative mood (mood-congruent), whereas people who scored higher on the PT subscale of empathy were more likely to listen to sad music when in a positive mood (mood-incongruent; Taruffi & Koelsch, 2014). The Fantasy component was positively associated with both mood-congruent and mood-incongruent liking of sad music. These findings propose that personality likely influences the situation in which people choose to engage with sad music by altering the emotional outcome of the music-listening experience (Garrido et al., 2017).

The current study

In the current study, we sought to clarify the relationship between individual differences, emotional responses, and the enjoyment of sad music. Previous literature has not statistically evaluated how personality influences the intensity and quality of emotional response to sad music and, subsequently, the motivations for engaging with sad music and in which situations. Here, we use regularized regression models with self-report measures of multiple dimensions of personality, including trait empathy and Absorption, the Big Five, and tendency to ruminate and reflect, to extend existing knowledge regarding the relationship between individual differences and our attraction to sad music. The aims of this study are to determine the combination of personality factors that are predictive of (a) the quality and intensity of emotional responses to sad music, (b) the situations in which people listen to sad music, and (c) the reasons and motivations people choose to engage with sad music. We then test whether emotional responses to sad music mediate the relationship between personality and the enjoyment of sad music in particular situations and for particular reasons.

We hypothesize that Fantasy and Absorption are correlated with intensity of positive feelings in response to sad music, whereas rumination is correlated with intensity of negative feelings. We predict that Fantasy and Absorption are linked to enjoying sad music because they trigger strong positive emotions, whereas rumination is linked to enjoying sad music because it prolongs negative emotions. Furthermore, we hypothesize that people who score high on Fantasy and Absorption report enjoying sad music when in aversive situations and that this relationship is mediated by positive feelings in response to sad music.

Methods

Participants

Participants were recruited from two separate sources: Amazon’s Mechanical Turk and the University of Southern California (USC) Undergraduate population. Mechanical Turk workers (n = 218, 117 female, M_age = 34.17, SD = 11.02) were paid three dollars for completing the survey and USC psychology students (n = 213, 174 female, M_age = 19.93, SD = 2.07) received course credit. All participants were living in the United States and spoke English as their primary language. English was the first language spoken by 88% of participants; 58% of participants identified as White, 6% as Hispanic, 6% as Black, 23% Asian/Pacific Islander, and 7% a mix of more than one cultural identity.

Materials

The online survey consisted of several personality questionnaires that were believed to be most relevant to the enjoyment of sad music: the 10-Item Personality Index (Gosling, Rentfrow, & Swann, 2003) for the Big Five, the Interpersonal Reactivity Index (Davis, 1983) for the four subscales of empathy, the Tellegen Absorption Scale (Tellegen & Atkinson, 1974), the Geneva Emotional Music Scale (GEMS-9; Zentner, Grandjean, & Scherer, 2008), the Rumination-Reflection Questionnaire (Trapnell & Campbell, 1999), the Southampton Nostalgia Scale (Barrett et al., 2010), and the Patient Health Questionnaire (Martin, Rief, Klaiberg, & Braehler, 2006) for severity of depressive symptoms. For brief descriptions of each, see the supplemental materials online.

In addition, participants were asked to rate how much they generally enjoyed listening to sad music. We chose to allow participants to think about pieces of music that evoke sadness regardless of how others may hear it (Van den Tol & Edwards, 2013) because we were not explicitly concerned with the characteristics that constitute a sad piece of music, but, rather, with how people feel emotional when listening to music that they believe is sad. The exact prompt can be found in the supplemental materials online.

Participants who reported enjoying sad music were then presented with a list of 12 different reasons or motivations for listening to sad music and were asked to rate how much they agreed or disagreed with each item using a 5-point Likert-type scale. The 12 items were adopted from Taruffi and Koelsch (2014) and can be found in the supplemental materials online.

In addition, the survey contained a list of 20 different situations in which a person might choose to listen to music and asked participants to describe the quality of the music that they would most likely listen to in that situation (see supplemental materials for full list of items). The items were adopted from previous studies that explored how music preferences change in various external situations (Thoma et al., 2012) and were selected to represent a range of both positive valence (e.g., “Describe the quality of the music that you would most likely listen to when you are celebrating a holiday, birthday, or other special occasion”) and negative valence (e.g., “Describe the quality of the music that you would most likely listen to when there has been a recent death or breakup”) circumstances. Using a slider from 0 (not at all) to 10 (very much), the participant answered three separate questions for each situation: (a) how happy sounding is the music; (b) how sad sounding is the music; (c) how energetic is the music. They were also given the option to select “Not Applicable.” The happiness and energy ratings were included for an additional research study not presented here.

Procedure

Following informed consent, participants were directed to a Qualtrics survey link with a unique, randomly chosen participant ID number. The questionnaires were presented randomly, with demographic information always assessed first.

Analysis

Exploratory factor analyses

Exploratory factor analyses (EFA) were conducted for the nine emotional responses to sad music (GEMS-9), the 20 situations for listening to sad music, and the 12 reasons for listening to sad music. For the list of 20 situations, only responses to the question “How sad sounding is the music” were used in the subsequent analyses. Scree plots were generated to select the number of underlying factors to extract based on an eigenvalue cutoff of one. Maximum-likelihood factor analysis with promax rotation was then conducted on all items from the survey using the R statistical program (R Development Core Team, 2017) factanal function in the stats package. Any item that loaded on more than one factor (with a factor loading score > 0.35) or failed to load on any factor (> 0.40) was removed and the factor analysis was conducted again. Regression scores from the final, pruned analyses were then extracted for subsequent analyses.

Lasso regression models

Separate lasso regression models were used to predict a) degree of enjoyment of sad music, b) emotional responses to sad music (GEMS-9 factors), and c) situational and motivational factor for listening to sad music based on all personality variables.

The lasso was chosen to determine the combination of personality variables that would best explain the enjoyment of sad music (Tibshirani, 1996). Predictor variables were selected using lasso regression on 80% of the data using 10-fold cross validation to tune the model and select the best-fitting penalty coefficient (lambda) based on the minimum error observed across folds. The final model was then validated on the left-out 20% of the data and model fit was assessed with mean squared error. Full regression models using the variables selected by the lasso regression were then conducted with all of the data to assess relationships between personality and (a) enjoying sad music, (b) emotional responses to sad music, (c) situations for listening to sad music, and (d) reasons for listening to sad music. All analyses were conducted using the R statistical program using the lme4 (multiple regression) and glmnet (lasso regression) packages.

Mediation models between personality and enjoying sad music

Mediation analyses were conducted to investigate the relationship between personality, emotional responses, and enjoying sad music in various situations and for various reasons. We used the results from the lasso regression models to guide these analyses. Specifically, we tested if emotional responses to sad music (as measured by GEMS-9) mediate the association between (a) personality and enjoying sad music, (b) personality and reasons for enjoying sad music, and (c) personality and listening to sad music in certain situations. Models were built using the lavaan package in R. For all models, items from the GEMS-9 that loaded onto a single factor were averaged and used as a single variable. Statistical code is made available on GitHub (https://github.com/sachsm53) and raw data are available upon request.

Results

Descriptive statistics

Summary statistics for all explanatory variables are presented in Table 1. A correlation table between all explanatory variables is presented in Supplemental Table 1 online. No significant differences in ratings of the enjoyment of sad music were found between two survey populations, M_Mturk = 3.31, M_USC = 3.37, t(429) = –0.60, p = .54, nor in ratings of how beautiful they found sad music, M_Mturk = 3.42, M_USC = 3.37, t(429) = –0.47, p = .63. Furthermore, both groups reported experiencing strong emotions in response to any type of music at similar frequencies, M_Mturk = 3.49, M_USC = 3.35, t(429) = –1.69, p = .09. Differences were found, however, in the number of hours per week that participants reported listening to sad music, M_Mturk = 4.06, M_USC = 3.08, t(429) = –3.07, p = .002. Given that the enjoyment of sad music question was the only question used as the dependent variable, we combined the responses from the two survey populations for all subsequent analyses.

Table 1.

Descriptive statistics of explanatory variables.

	n	M	SD	Min	Pctl (25)	Pctl (75)	Max
Years of Music Training	341	7.35	5.23	1	4.00	10	22
Age	431	27.13	10.68	17	19	32	77
Extraversion	431	3.94	1.69	1	2.5	5.5	7
Agreeability	431	5.13	1.23	1	4.5	6	7
Conscientiousness	431	5.44	1.17	2	4.50	6.50	7
Emotional Stability	431	4.57	1.51	1	3.5	6	7
Openness to experience	431	4.14	0.93	2	3.5	4.5	7
Depression severity	431	6.46	5.69	0	2	9	27
Fantasy	431	2.65	0.75	0	2.1	3.1	4
Perspective taking	431	2.66	0.69	0	2.29	3.14	4
Empathic concern	431	2.89	0.68	0	2.43	3.43	4
Personal distress	431	1.76	0.77	0	1.3	2.3	4
Absorption	431	3.55	0.70	1.25	3.17	4.00	5
Rumination	431	3.44	0.85	1	2.92	4.00	5
Reflection	431	3.45	0.82	1	2.92	4.00	5
Nostalgia	431	4.61	1.50	1	3.75	5.75	7

Underlying emotions in response to sad music

An exploratory factor analysis suggested a three-factor model for the GEMS-9. All nine items were kept in the final model (see Supplemental Table 2 online). The solution replicated the three-factor model found in Zentner et al. (2008); we therefore refer to the three factors by the names used in Zentner et al. (2008): sublime (Cronbach’s α = .79), vital (Cronbach’s α = .76), and unease (Cronbach’s α = .38).

Predicting the enjoyment of sad music and emotional responses to sad music

All 17 variables were included in the model to predict ratings for the enjoyment of sad music. The final Lasso model suggested six predictor variables (see Table 2). In the full regression model, Fantasy positively correlated with enjoying sad music (β = .17, SE = 0.06, p = .003) as was Absorption (β = .16, SE = 0.05, p = .002).

Table 2.

Final lasso model predicting the enjoyment of sad music and emotional responses to sad music (factors of the GEMS-9).

	Enjoy sad music	Sublime Factor	Vital Factor	Unease Factor
Intercept	−0.00 (0.05)	−0.03 (0.04)	0.04 (0.05)	0.02 (0.04)
Gender	–	−0.08 (0.05)	0.10* (0.05)	0.05 (0.04)
Years of Music Training	–	–	0.06 (0.04)	–
Age	–	−0.08 (0.04)	−0.05 (0.05)	–
Extraversion	−0.13** (0.05)	−0.10* (0.04)	0.09 (0.04)	0.08* (0.04)
Agreeability	–	0.08 (0.05)	−0.11* (0.05)	–
Conscientiousness	–	–	−0.08 (0.05)	−0.03 (0.04)
Emotional Stability	–	0.15** (0.05)	0.07 (0.06)	−0.04 (0.05)
Openness to experience	–	–	0.08 (0.04)	0.03 (0.03)
Depression severity	0.07 (0.05)	–	0.02 (0.05)	0.06 (0.04)
Fantasy	0.17** (0.06)	0.15** (0.05)	−0.11* (0.05)	–
Perspective taking	–	–	–	−0.06 (0.04)
Empathic concern	−0.10 (0.05)	–	–	0.09* (0.04)
Personal distress	–	–	–	0.09* (0.04)
Absorption	0.16** (0.05)	0.15** (0.05)	−0.01 (0.05)	0.09* (0.04)
Rumination	0.07 (0.05)	0.12* (0.06)	–	0.03 (0.05)
Reflection	–	0.04 (0.05)	–	–
Nostalgia	–	0.14** (0.05)	–	0.05 (0.04)
Observations	431	431	431	431
R ²	.14	.22	.08	.12
Adjusted R²	.12	.20	.06	.10
Residual SE	0.94 (df = 424)	0.86 (df = 420)	0.91 (df = 419)	0.70 (df = 418)
F statistic	11.06*** (df = 6, 424)	11.61*** (df = 10, 420)	3.54*** (df = 11, 419)	4.77*** (df = 12, 418)

GEMS-9: Geneva Emotional Music Scale; SE: standard error.

p < .05. **p < .01. ***p < .001.

When predicting factor scores from the sublime factor of the GEMS-9, the best-fitting lasso regression retained 10 predictor variables (see Table 2). In the full multiple regression model, emotional stability (β = .15, SE = 0.04, p = .005), Fantasy (β = .15, SE = 0.05, p = .003), Absorption (β = .15, SE = 0.05, p = .006), rumination (β = .12, SE = 0.06, p = .04), and nostalgia proneness (β = .14, SE = 0.05, p = .003) were positively correlated with sublime feelings, whereas extraversion was negatively correlated with sublime feelings (β = –.10, SE = 0.04, p = .02)

The best-fitting lasso regression predicting the vital factor retained 11 predictor variables. Gender (β = –.21, SE = 0.10, p = .04), agreeability (β = –.11, SE = 0.05, p = .02), and fantasy (β = –.10, SE = 0.05, p = .04) were negatively correlated with the vital factor. Finally, predicting unease factor, the best-fitting lasso model retained 12 predictor variables. The unease factor was positively correlated with extraversion (β = .08, SE = 0.04, p = .02), EC (β = .09, SE = 0.04, p = .04), PD (β = .09, SE = 0.04, p = .04), and Absorption (β = .09, SE = 0.04, p = .02).

Situations in which one would listen to sad music

The scree-plot suggested a four-factor model adequately accounted for the variance in the 20 possible situations. The first four factors had an eigenvalue greater than 1. One item, when wanting to get-away did not have a loading score above 0.40 on any of the four factors and was removed from the model. In the final, pruned model, the four factors explained 45% of the total variance in the data and intra-factor consistencies were high (see Table 3). Factor 1 included situations that were positive, yet solitary, such as being creative or trying to focus and was therefore labeled “positive self.” Factor 2 corresponded to situations marked by sadness and nostalgia and was therefore labeled “melancholy.” Factor 3 corresponded to situations that were stressful and anxious, and was therefore labeled “tension.” Finally, Factor 4 corresponded to situations with positive valence as well as a social component, and was therefore labeled “positive other.”

Table 3.

EFA with promax-rotated loadings for 20 situations on four factors.

Situations	F1: Positive self	F2: Melancholy	F3: Tension	F4: Positive other
Feeling relaxed	0.78
Doing something else, driving, traveling, reading, working	0.64
You are alone	0.63
Being creative or feeling inspired	0.62
In contact with nature	0.61
You are happy or in a good mood	0.57
You are trying to focus	0.53
You are missing someone		0.84
You are sad or in a bad mood		0.74
There has been a recent death or breakup		0.64
Feeling lonely		0.61
Feeling homesick		0.56
Feeling nostalgic or reflecting on past persona experiences		0.53
Feeling frustrated			0.79
Feeling angry			0.70
Feeling stressed			0.61
You are celebrating a holiday, birthday, or other social occasion				0.77
You are with friends/at a social gathering				0.67
You are running, dancing, working out				0.56
You want to distance yourself or get away from your problems
Loading	2.96	2.81	1.59	1.40
Variance explained	16%	15%	8%	7%
Cronbach’s α	.83	.84	.78	.72

N = 431. Factor loadings from pre-pruned model < .|40| omitted. EFA: exploratory factor analysis.

Lasso regression was then used to determine the combination of personality factors that predict loading scores from each of these four situational factors (see Table 4). In the final regression model predicting the positive self factor, reflection (β = .19, SE = 0.06, p = .001) was positively correlated with factor scores and EC (β = –.15, SE = 0.04, p = .03) was negatively correlated with factor scores. In the best-fitting regression model predicting the melancholy factor, only Fantasy was positively correlated with factor scores (β = .30, SE = 0.05, p < .001). For the tension factor, rumination was found to be positively correlated with factor scores (β = .14, SE = 0.06, p = .02). For the positive other factor, gender (β = .15, SE = 0.05, p = .005), years of musical training (β = .10, SE = 0.05, p = .04), and Openness to Experience (β = .10, SE = 0.05, p = .04) were positively correlated with factor scores, whereas conscientiousness (β = –.10, SE = 0.04, p = .04), emotional stability (β = –.16, SE = 0.07, p = .03), and rumination (β = –.22, SE = 0.06, p < .001) were negatively correlated with factor scores.

Table 4.

Final lasso model predicting factor scores from situational factors for listening to sad music.

	F1: Positive Self	F2: Melancholy	F3: Tension	F4: Positive Other
Intercept	0.00 (0.06)	−0.00 (0.05)	−0.03 (0.09)	0.21* (0.09)
Gender	–	–	0.04 (0.12)	−0.31** (0.11)
Years of Music Training	–	–	–	0.010 (0.05)
Age	–	–	−0.07 (0.05)	0.09 (0.05)
Extraversion	–	–	–	–
Agreeability	−0.07 (0.07)	–	–	0.05 (0.06)
Conscientiousness	–	–	–	−0.10* (0.05)
Emotional Stability	–	–	–	−0.16* (0.07)
Openness to experience	–	–	–	0.10* (0.05)
Depression severity	0.08 (0.07)	–	–	0.003 (0.06)
Fantasy	–	0.30*** (0.05)	–	0.09 (0.05)
Perspective taking	–	–	–	–
Empathic concern	−0.15* (0.07)	–	–	−0.10 (0.06)
Personal distress	–	–	–	0.08 (0.06)
Absorption	–	–	−0.06 (0.06)	−0.07 (0.06)
Rumination	0.04 (0.07)	–	0.14* (0.06)	−0.22*** (0.06)
Reflection	0.19** (0.06)	–	−0.11 (0.06)	–
Nostalgia	–	–	–	–
Observations	431	431	431	431
R ²	.05	.07	.04	.11
Adjusted R²	.04	.07	.03	.08
Residual SE	1.19	1.06	1.10	0.96
F statistic	4.59*** (df = 5, 425)	33.93*** (df = 1, 429)	3.23** (df = 5, 425)	3.93*** (df = 13, 417)

SE: standard error.

p < .05. **p < .01. ***p < .001.

Reasons for enjoying sad music

Based on a scree-plot and the eigenvalue criteria, it was determined that a three-factor model was appropriate to explain the variance in response to the 12 possible reasons for enjoying sad music. Three items did not have a factor loading score above 0.40 and were not included in the final model. In the final, pruned model, the items that loaded on Factor 1 related to using sad music to regulate positive emotional responses (see Table 5). The items that loaded on the second factor express benefits of sad music related to the engagement of cognitive processes to better resonate with one’s own emotions and the emotions of others. The items that loaded on Factor 3 relate to the idea that sadness-induced through music can be beneficial by embracing negative emotions, through the process of prolonging or purging them.

Table 5.

Final, pruned EFA with promax-rotated loadings for nine reasons for listening to sad music on three factors.

Reasons	F1:Positive emotion regulation	F2:Cognitive empathy	F3:Negative emotion regulation
I am more likely to feel touched, awed, or moved by sad music	0.73
It generally results in a strengthening and/or deepening of emotions and I find powerful emotions enjoyable	0.63
It makes me feel more calm, soothed, and/or relaxed than other types of music	0.60
It helps me regulate my mood and/or emotions better than other types of music	0.42
It makes me think more realistically about my life and current situation		0.80
It helps me gain a better understanding of my own feelings or my current situation		0.65
It allows me to relate to and/or feel connected to others		0.38
It makes me feel more sad and/or prolongs my sadness and I think feeling truly sad has some value or importance			0.62
It allows me to release or purge myself of negative emotions			0.40
Listening to sad music is a way for me to express how I am feeling and/or who I am
It is more likely to trigger an emotional response, which reassures me of my ability to feel emotions at all
I can experience a milder version of the emotions expressed in the music without the negative life consequences
Loading	1.48	1.38	0.74
Variance explained	16%	15%	8%
Cronbach’s α	.71	.65	.42

N = 431. Factor loadings from pre-pruned model < .|40| omitted. EFA: exploratory factor analysis.

In the regression predicting Factor 1 (positive emotion regulation), Absorption (β = .15, SE = 0.06, p = .02) and rumination (β = .14, SE = 0.05, p = .02) were positively correlated with factor scores. When predicting Factor 2 (cognitive empathy), PT was positively correlated with factor scores (β = .17, SE = 0.05, p < .001). For Factor 3 (negative emotion regulation), Fantasy (β = .10, SE = 0.05, p = .04) and rumination (β = .10, SE = 0.05, p = .05) were positively correlated with factor scores (see Table 6).

Table 6.

Final lasso model predicting factor scores from the reasons for listening to sad music.

	F1: Positive emotion regulation	F2: Cognitive empathy	F3: Negative emotion regulation
Intercept	−0.03 (0.05)	0.02 (0.10)	−0.01 (0.04)
Gender	–	−0.05 (0.12)	–
Years of Music Training	–	–	–
Age	–	–	–
Extraversion	–	–	–
Agreeability	–	–	–
Conscientiousness	–	–	–
Emotional Stability	–	–	−0.09 (0.05)
Openness to experience	–	0.10 (0.06)	–
Depression severity	–	–	–
Fantasy	–	–	0.10* (0.05)
Perspective taking	–	0.20*** (0.06)	–
Empathic concern	–	–	−0.002 (0.04)
Personal distress	–	–	–
Absorption	0.14* (0.06)	0.07 (0.06)	–
Rumination	0.15** (0.06)	–	0.10* (0.05)
Reflection	–	–	–
Nostalgia	0.07 (0.06)	0.04 (0.06)	–
Observations	332	332	332
R ²	.07	.07	.07
Adjusted R²	.06	.05	.06
Residual SE	0.95	0.99	0.75
F statistic	7.99*** (df = 3, 328)	4.57*** (df = 5, 326)	5.98*** (df = 4, 327)

SE: standard error.

p < .05. **p < .01. ***p < .001.

The mediating effect of emotions

We next evaluated whether emotional responses to sad music mediate the relationship between personality measures and enjoying sad music. Because the multiple regression results showed that Fantasy, rumination, and Absorption were all positively correlated with enjoying sad music, we chose to focus on these three personality measures to further tease apart their relationship with response to sad music. Additional models are presented in the supplemental materials.

We first tested whether the relationship between fantasy and enjoying sad music was mediated by an increase in positive emotions. Significant indirect and direct effects were found between enjoyment of sad music and Fantasy with sublime feelings as a mediator, β_indirect = .09, SE = 0.02, p < .001; β_direct = .15, SE = 0.05, p = .002; see Figure 1(a), suggesting partial mediation. Significant indirect and direct effects were also found with Absorption as the predictor variable, β_indirect = .10, SE = 0.02, p < .001; β_direct = .16, SE = 0.05, p = .001; see Figure 1(b).

Figure 1.

Mediation Models With Emotional Responses to Sad Music Mediating the Relationship Between Personality and the Enjoyment of Sad Music. (a) Sublime Emotions in Response to Sad Music Was Found to Partially Mediate the Relationship Between Fantasy and Enjoying Sad Music. (b). Sublime Emotions in Response to Sad Music Was Found to Partially Mediate the Relationship Between Absorption and Enjoying Sad Music.

We then evaluated whether negative feelings of unease mediate the relationship between rumination and enjoying sad music. Contrary to our hypothesis, a negative indirect effect with a positive direct effect was found (β_indirect = –.04, SE = 0.01, p = .003; β_direct = .23, SE = 0.05, p < .001), suggesting that rumination is directly related to enjoying sad music, but negatively related to sad music through its positive associations with feelings of unease (β = .17, SE = 0.04). On the contrary, a significant positive indirect and direct effect was found with sublime feelings mediating the relationship between rumination and enjoying sad music (β_indirect = .07, SE = 0.02, p < .001; β_direct = .13, SE = 0.05, p = .007, see Figure 2).

Figure 2.

Mediation Model With Emotional Responses to Sad Music Mediating the Relationship Between Rumination and the Enjoyment of Sad Music. Unease Emotions in Response to Sad Music Had a Negative Indirect Effect on the Relationship Between Rumination and Enjoying Sad Music. Sublime Emotions in Response to Sad Music Were Found to Partially Mediate the Relationship Between Rumination and Enjoying Sad Music.

To further understand this relationship, we conducted mediation models to determine if negative (unease) and positive (sublime) feelings mediate the relationship between personality and reasons for enjoying sad music. We focused on the items of the third factor, related to using sad music to regulate negative emotions. Given that the reliability scores for this third factor were low, we decided to build two separate models, one that predicted responses to the purge item and one that predicts responses to the prolong item.

With rumination predicting enjoyment because it prolongs feelings of sadness, a significant indirect and direct effect was found with feelings of unease as a mediator (β_indirect = .07, SE = 0.02, p < .001; β_direct = .20, SE = 0.06, p = .001). With rumination predicting purging negative emotions, significant positive indirect effects were found with both unease feelings as a mediator (β_indirect = .05, SE = 0.02, p = .004; β_direct = .03, SE = 0.06, p = .62) and with sublime feelings as a mediator (β_indirect = .03, SE = 0.02, p = .03; β_direct = .06, SE = 0.06, p = .34; see Figure 3).

Figure 3.

Mediation Model With Unease Emotions Mediating the Relationship Between Rumination and Reasons for Enjoying Sad Music. Unease Emotions and sublime Emotions in Response to Sad Music Was Found to Mediate the Relationship Between Rumination and Enjoying Sad Music Because It Purges Negative Feelings.

Opposite effects were found with Fantasy as the predictor variable: no indirect effect was found between enjoying sad music because it purges negative emotions through sublime feelings (β_indirect = .02, SE = 0.01, p = .09), whereas a small, indirect effect was found when mediated by unease (β_indirect = .03, SE = 0.01, p = .02; β_direct = .21, SE = 0.06, p < .001). On the contrary, a significant indirect effect was found between Fantasy and enjoyment because it prolongs feelings of sadness with feelings of unease as the mediator, β_indirect = .06, SE = 0.02, p = .001; β_direct = .07, SE = 0.06, p = .24; see Figure 4(a). Nearly identical results were found when using Absorption as the explanatory variable, rather than Fantasy, to predict enjoying sad music because it prolongs feelings of sadness, β_indirect = .07, SE = 0.02, p = .003; β_direct = .06, SE = 0.06, p = .33, see Figure 4(b).

Figure 4.

Mediation Model With Unease Emotions Mediating the Relationship Between Fantasy/Absorption and Reasons for Enjoying Sad Music. (a). Unease Emotions in Response to Sad Music Was Found to Mediate the Relationship Between Fantasy and Enjoying Sad Music Because It Prolongs Feeling of Sadness. (b). Unease emotions in Response to Sad Music Was Additionally Found to Mediate the Relationship Between Absorption and Enjoying Sad Music Because It Prolongs Feeling of Sadness. No Effect Was Found With Sublime Feelings as a Mediator.

Finally, we evaluated the mediating effect of emotions on the relationship between personality and the situations in which people listen to sad music. We focused on predicting the melancholy factor, given its high internal reliability and its relevancy to the topic of sad music. A positive indirect effect was found with sublime feelings mediating the relationship between Absorption and listening to sad music in melancholic situations (β_indirect = .14, SE = 0.03, p < .001; β_direct = .08, SE = 0.05, p = .24). In addition, a significant indirect and direct effect was with sublime feelings meditating the relationship between Fantasy and listening to sad music in sad situations (β_indirect = .11, SE = 0.02, p < .001; β_direct = .19, SE = 0.05, p < .001). Significant indirect effects were not found when using unease feelings as the mediating factor (β_indirect = .008, SE = 0.01, p = .66, for Absorption; β_indirect = .006, SE = 0.01, p = .36, for Fantasy). When using rumination as the predictor variable, a significant indirect and direct effect was found with sublime emotions (β_indirect = .08, SE = 0.02, p = .001; β_direct = .12, SE = 0.05, p = .01), but not with unease emotions (β_indirect = .009, SE = 0.01, p = .89).

Discussion

The findings from this study support those of previous investigations by showing that the enjoyment of sad music is related to specific personality measures. They additionally extend past results by highlighting how these relationships depend on (a) situational factors, (b) individual reasons for engaging with sad music, and (c) emotional responses to sad music. When controlling for other personality measures, the Fantasy subcomponent of empathy, and a related measure—trait Absorption—uniquely predicted the enjoyment of sad music. Moreover, both personality measures were correlated with sublime emotions in response to sad music, as measured by the GEMS-9. Mediation models additionally confirmed that Fantasy and Absorption are indirectly positively correlated with enjoying sad music through the activation of sublime feelings. Overall, these results provide support for previous hypotheses regarding the relationship between Fantasy, Absorption, and the enjoyment of sad music; they show that the act of becoming engrossed and immersed in works of art that convey sadness engenders strong, positive emotions rather than negative ones.

By including questions regarding individual’s reasons for engaging with sad music, we then clarify the rewarding aspects of sad music in high Fantasy and high Absorption participants. When controlling for additional personality factors, we show that trait Absorption is associated with enjoying sad music because of it produces and regulates strong, positive feelings, whereas Fantasy is associated with enjoying music because it can purge and/or prolong negative feelings. Although Fantasy and Absorption are highly related traits, our results suggest that each might be associated with the enjoyment of sad music through separate mechanisms. Given that Fantasy relates to a tendency to become engaged and transported into narratives, whereas Absorption is far more general, it may be that fantasy-seekers are more likely to resonate specifically with the emotions conveyed through a piece of sad art, thus enabling them to purge and/or prolong negative feelings. This is in line with the results from a previous music-listening study, in which it was shown that Fantasy, but not Absorption, was correlated with being moved by unfamiliar sad music (Eerola, Vuoskoski, & Kautiainen, 2016).

Rumination was also found to be positively associated with sublime feelings in response to sad music and not, as predicted, with feelings of unease. Interestingly, rumination was associated with enjoying sad music for reasons related to the strengthening of positive emotions, calling into question the hypothesis proposed by Garrido and Schubert (2013), that people who tend to ruminate have a “maladaptive attraction” to sad music. Other studies have also shown both rumination and Fantasy/Absorption can be independently related to liking sad music through separate mechanisms (Schubert et al., 2018). We extend these findings and clarify this link by testing separate mediation models with motivational factors. The results showed that the correlation between rumination and enjoying sad music because it prolongs negative emotions was partially mediated by feelings of unease, providing some evidence for an unhealthy response to sad music. On the contrary, a relationship between rumination and enjoying sad music because it purges negative feelings was fully mediated by both feelings of sublime and unease. Therefore, it may be that people who tend to ruminate can at times feel positive emotions and at other times negative emotions in response to sad music, each serving a different function that may not be exclusively “maladaptive” or “adaptive.”

In addition, our results show that the situations in which people choose to engage with sad music is correlated with unique personality profiles. People who score higher on Fantasy were more likely to listen to sad music in melancholic situations, such as after a breakup or when homesick. We further tested whether this relationship was mediated by sublime feelings and report evidence for a partial mediation effect. This suggests that people who tend to transport themselves imaginatively into works of art can experience intense aesthetic emotions when listening to sad music, which may help with coping or overcoming their current negative circumstances.

Openness to Experience was positively correlated with listening to sad music in positive, social situations, such as with friends and when celebrating. Previous studies have shown that people who score higher in Openness to Experience are more likely to feel intense emotions to music (Liljestrom, Juslin, & Vastfjall, 2013; Nusbaum & Silvia, 2011) and are more comfortable with mixed emotional experiences, that is, simultaneous feelings of happiness and sadness (Barford & Smillie, 2016). It is possible that people who express Openness to Experience enjoy listening to sad music in social situations because they appreciate this mixture of feelings.

With regard to EC, we replicated previous findings in showing that this trait was positively correlated with negative feelings in response to sad music (Taruffi & Koelsch, 2014). Furthermore, we showed that EC was negatively correlated with listening to sad music in positive, self-focused situations, such as when trying to relax or in contact with nature. These results suggest that this tendency to feel concern for the pain of others triggers negative feelings in response to sad music that are not particularly beneficial, making it less likely that such individuals would choose to listen to such music when trying to maintain a positive mood.

We did not find a positive relationship between enjoying sad music and PT, which has been shown in some previous studies (Kawakami & Katahira, 2015; Taruffi & Koelsch, 2014). However, we did find that individuals who scored higher in PT reported enjoying sad music because it conferred psychological benefits related to understanding one’s emotions and the emotions of others. Our results provide one possible explanation for inconsistent findings with regard to the link between PT and enjoying sad music; that people with an ability to adopt the psychological perspectives of others can find sad music enjoyable when it enables them to process their own feelings through the act of connecting with the feelings of others.

It is important to note that the results here are based solely on self-report measures. Previous results have suggested that direct measures of affective responses to music in individuals prone to depression or rumination do not always align with indirect measures (McFerran & Saarikallio, 2014). There is evidence that, while people may expect sad music to repair their mood, in actuality, such strategies are unsuccessful at times, leading to a worse mood (Van den Tol & Edwards, 2013). It is also possible that in certain patient populations, sad music can become linked with a perverse enjoyment of negative experiences, but that this “reward” is not necessarily beneficial. Further research with more objective measures of affective response are therefore needed to corroborate the link between rumination, music-evoked pleasure, and salubrious outcomes.

Another limitation of the current study is the use of the GEMS-9 as the measure of emotional responses to sad music. We chose this scale because it has been validated in previous literature (Zentner et al., 2008) and has been used by numerous studies exploring music-evoked emotions (Eerola & Vuoskoski, 2010; Taruffi & Koelsch, 2014). However, the items tend to skew toward the positive (only two of the nine items relate specifically to sadness). It may well be, then, that the GEMS is not well-suited to capture the complexity and nuances of the feelings that result from listening to sad music. Indeed, a recent study found evidence for three qualitatively distinct experiences of music-evoked sadness (Peltola & Eerola, 2016), none of which were captured by the items of the GEMS-9 (Eerola & Peltola, 2016). Future investigations of the enjoyment of sad music would benefit from incorporating a more multifaceted model of music-evoked sadness.

In sum, our findings suggest that personality traits are linked to the enjoyment of sad music through separate mechanisms. These mechanisms partially depend on one’s emotional responses to sad music. The results therefore emphasize the complex relationship between individual differences, learned associations, and the external environment that leads to the appreciation of music that conveys negative emotions. Clarifying this relationship and being able to predict when and how certain stimuli that express negative affect will produce pleasure will help expand the role that music can play in everyday life and in therapeutic settings.

Supplemental Material

Supplemental_material – Supplemental material for Unique personality profiles predict when and why sad music is enjoyed

Supplemental material, Supplemental_material for Unique personality profiles predict when and why sad music is enjoyed by Matthew E Sachs, Antonio Damasio and Assal Habibi in Psychology of Music

Footnotes

Authors’ note

Matthew Sachs is also affiliated with the Department of Psychology, Columbia University, New York, NY, USA.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Matthew E Sachs

Supplemental material

Supplemental material for this article is available online.

References

Barford

K. A.

Smillie

L. D.

(2016). Openness and other Big Five traits in relation to dispositional mixed emotions. Personality and Individual Differences, 102, 118–122. doi:10.1016/j.paid.2016.07.002

Barrett

F. S.

Grimm

K. J.

Robins

R. W.

Wildschut

Sedikides

Janata

(2010). Music-evoked nostalgia: Affect, memory, and personality. Emotion, 10, 390–403. doi:10.1037/a0019006

Damasio

Carvalho

G. B.

(2013). The nature of feelings: Evolutionary and neurobiological origins. Nature Reviews Neuroscience, 14, 143–152. doi:10.1038/nrn3403

Davis

M. H.

(1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology, 44, 113–126. doi:10.1037/0022-3514.44.1.113

Eerola

Peltola

H.-R.

(2016). Memorable experiences with sad music—Reasons, reactions and mechanisms of three types of experiences. PLoS ONE, 11, Article e0157444. doi:10.1371/journal.pone.0157444

Eerola

Vuoskoski

J. K.

(2010). A comparison of the discrete and dimensional models of emotion in music. Psychology of Music, 39, 18–49. doi:10.1177/0305735610362821

Eerola

Vuoskoski

J. K.

Kautiainen

(2016). Being moved by unfamiliar sad music is associated with high empathy. Frontiers in Psychology, 7, Article 1176. doi:10.3389/fpsyg.2016.01176

Egermann

Kopiez

Altenmüller

(2013). The influence of social normative and informational feedback on musically induced emotions in an online music listening setting. Psychomusicology: Music, Mind, and Brain, 23, 21–32. doi:10.1037/a0032316

Egermann

Sutherland

M. E.

Grewe

Nagel

Kopiez

Altenmuller

(2011). Does music listening in a social context alter experience? A physiological and psychological perspective on emotion. Musicae Scientiae, 15, 307–323. doi:10.1177/1029864911399497

10.

Forgas

J. P.

(2013). Don’t worry, be sad! On the cognitive, motivational, and interpersonal benefits of negative mood. Current Directions in Psychological Science, 22, 225–232. doi:10.1177/0963721412474458

11.

Garrido

Eerola

McFerran

(2017). Group rumination: Social interactions around music in people with depression. Frontiers in Psychology, 8, Article 490. doi:10.3389/fpsyg.2017.00490

12.

Garrido

Schubert

(2011). Negative Emotion in Music: What is the Attraction? A Qualitative Study. Empirical Musicology Review, 6(4), 214–230. https://doi.org/10.18061/1811/52950

13.

Garrido

Schubert

(2013). Adaptive and maladaptive attraction to negative emotions in music. Musicae Scientiae, 17, 147–166. doi:10.1177/1029864913478305

14.

Garrido

Schubert

(2015). Moody melodies: Do they cheer us up? A study of the effect of sad music on mood. Psychology of Music, 43, 244–261. doi:10.1177/0305735613501938

15.

Gosling

S. D.

Rentfrow

P. J.

Swann

W. B.

(2003). A very brief measure of the Big-Five personality domains. Journal of Research in Personality, 37, 504–528. doi:10.1016/S0092-6566(03)00046-1

16.

Hogue

J. D.

Crimmins

A. M.

Kahn

(2016). “So sad and slow, so why can’t I turn off the radio”: The effects of gender, depression, and absorption on liking music that induces sadness and music that induces happiness. Psychology of Music, 44, 816–829. doi:10.1177/0305735615594489

17.

Juslin

P. N.

Liljeström

Västfjäll

Barradas

Silva

(2008). An experience sampling study of emotional reactions to music: Listener, music, and situation. Emotion, 8, 668–683. doi:10.1037/a0013505

18.

Kawakami

Katahira

(2015). Influence of trait empathy on the emotion evoked by sad music and on the preference for it. Frontiers in Psychology, 6, Article 1541. doi:10.3389/fpsyg.2015.01541

19.

Kivy

(1991). Music alone: Philosophical reflections on the purely musical experience. Ithaca, NY: Cornell University Press.

20.

Ladinig

Schellenberg

E. G.

(2012). Liking unfamiliar music: Effects of felt emotion and individual differences. Psychology of Aesthetics, Creativity, and the Arts, 6, 146–154. doi:10.1037/a0024671

21.

Levinson

(1997). Music and negative emotions. In Robinson

(Ed.), Music and meaning (pp. 215–241). Ithaca, NY: Cornell University Press.

22.

Liljestrom

Juslin

P. N.

Vastfjall

(2013). Experimental evidence of the roles of music choice, social context, and listener personality in emotional reactions to music. Psychology of Music, 41, 579–599. https://doi.org/10.1177/0305735612440615

23.

Martin

Rief

Klaiberg

Braehler

(2006). Validity of the Brief Patient Health Questionnaire Mood Scale (PHQ-9) in the general population. General Hospital Psychiatry, 28, 71–77. doi:10.1016/j.genhosppsych.2005.07.003

24.

McCrae

R. R.

(2007). Aesthetic chills as a universal marker of openness to experience. Motivation and Emotion, 31(1), 5–11. doi:10.1007/s11031-007-9053-1

25.

McFerran

K. S.

Saarikallio

(2014). Depending on music to feel better: Being conscious of responsibility when appropriating the power of music. Arts in Psychotherapy, 41, 89–97. doi:10.1016/j.aip.2013.11.007

26.

Moore

M. T.

Fresco

D. M.

(2012). Depressive realism: A meta-analytic review. Clinical Psychology Review, 32, 496–509. doi:10.1016/j.cpr.2012.05.004

27.

Nusbaum

E. C.

Silvia

P. J.

(2011). Shivers and timbres: Personality and the experience of chills from music. Social Psychological and Personality Science, 2, 199–204. doi:10.1177/1948550610386810

28.

Oliver

M. B.

(1993). Exploring the paradox of the enjoyment of sad films. Human Communication Research, 19, 315–342. Retrieved from https://broadly.vice.com/en_us/article/j5vvex/exploring-the-paradox-of-trans-visibility-trap-door

29.

Peltola

Eerola

(2016). Fifty shades of blue: Classification of music-evoked sadness. Musicae Scientiae, 20(1), 84–102. doi:10.1177/1029864915611206

30.

R Development Core Team. (2017). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.

31.

Robinson

(1994). The expression and arousal of emotion in music. The Journal of Aesthetics and Art Criticism, 52(1), 13–22.

32.

Sachs

M. E.

Damasio

Habibi

(2015). The pleasures of sad music: A systematic review. Frontiers in Human Neuroscience, 9, Article 404. doi:10.3389/fnhum.2015.00404

33.

Schubert

Halpern

A. R.

Kreutz

Garrido

(2018). Attraction to sad music: The role of imagery, absorption, and rumination. Psychology of Aesthetics, Creativity, and the Arts, 12, 251–258. doi:10.1037/aca0000160

34.

Taruffi

Koelsch

(2014). The paradox of music-evoked sadness: An online survey. PLoS ONE, 9(10), Article e110490. doi:10.1371/journal.pone.0110490

35.

Thoma

M. V.

Ryf

Mohiyeddini

Ehlert

Nater

U. M.

(2012). Emotion regulation through listening to music in everyday situations. Cognition and Emotion, 26(3), 550–560. https://doi.org/10.1080/02699931.2011.595390

36.

Tellegen

Atkinson

(1974). Openness to absorbing and self-altering experiences (‘absorption’), a trait related to hypnotic susceptibility. Journal of Abnormal Psychology, 83, 268–77.

37.

Tibshirani

(1996). Regression shrinkage and selection via the lasso. Royal Statistical Society, 58, 267–288.

38.

Tellegen

Atkinson

(1974). Openness to absorbing and self-altering experiences (‘absorption’), a trait related to hypnotic susceptibility. Journal of Abnormal Psychology, 83, 268–277.

39.

Thoma

M. V.

Ryf

Mohiyeddini

Ehlert

Nater

U. M.

(2012). Emotion regulation through listening to music in everyday situations. Cognition and Emotion, 26(3), 550–560. doi:10.1080/02699931.2011.595390

40.

Trapnell

P. D.

Campbell

J. D.

(1999). Private self-consciousness and the five-factor model of personality: Distinguishing rumination from reflection. Journal of Personality and Social Psychology, 76(2), 284–304.

41.

Van den Tol

J. M.

Edwards

. (2013). Exploring a rationale for choosing to listen to sad music when feeling sad. Psychology of Music, 41, 440–465. https://doi.org/10.1177/0305735611430433

42.

Vuoskoski

Thompson

W. F.

McIlwain

Eerola

(2011). Who enjoys listening to sad music and why? Music Perception, 29, 311–318.

43.

Wilhelm

Gillis

Schubert

Whittle

E. L.

(2013). On a blue note: Depressed peoples’ reasons for listening to music. Music and Medicine, 5(2), 76–83. doi:10.1177/1943862113482143

44.

Zentner

Grandjean

Scherer

K. R.

(2008). Emotions evoked by the sound of music: Characterization, classification, and measurement. Emotion, 8, 494–521. doi:10.1037/1528-3542.8.4.494

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.65 MB