Energy,Popularity,and the Circumplex

Abstract

The circumplex model of affect claims that emotions can be understood in terms of their relative positions along two dimensions, namely pleasant-unpleasant and active-sleepy; and numerous studies of small samples of music have yielded data consistent with this. The present research tests whether the energy and beats per minute (proxies for the arousal dimension) and popularity as expressed in terms of sale charts (a possible proxy for the pleasantness dimension) could predict scores on six moods in 143,353 musical pieces. Findings concerning energy were clearly consistent with the circumplex model; findings for beats per minute were consistent though more equivocal; and findings concerning popularity yielded only limited support. Numerous relationships between popularity and mood were indicative of the commercial market for specific genres; and evidence demonstrated considerable differences in the mood scores between genres. In addition to the circumplex model and aesthetic responses, the findings have implications for music marketing, therapy, and everyday listening.

Keywords

music emotion circumplex popularity sales

Many attempts to understand emotion in music have done so by considering the degree of activity in the music stimuli. North and Hargreaves (2008) and Sloboda and Juslin (2001) review numerous attempts in which participants have been typically asked to assess target pieces in terms of concepts such as arousal, orderliness, complexity, or energy, and these assessments are then mapped onto assessments of the more fine-grained details of emotional responses to those pieces. While many of these attempts have been successful, their obvious limitation is that they have employed a relatively narrow range of musical stimuli, which are often composed specifically for the research in question and presented to undergraduate participants under laboratory conditions. In contrast, the present research attempts to determine whether the activity of commercially successful pieces of music can predict their emotional connotations across 143,353 unique pieces, which in effect represent the entire corpus of music that has enjoyed any degree of commercial success in the United Kingdom.

Sloboda and Juslin (2001) outline three major psychological approaches to conceptualizing emotion, namely categorical, prototype, and dimensional. The first of these, the categorical approach, argues that more complex emotions are developed through the amalgamation of clearly distinguishable “basic emotions” (such as fear or happiness), which are themselves of adaptive significance. In contrast, within the prototype approach, emotions are structured in a hierarchy in which a given specific emotion is related less or more closely to the more general emotion located in the superordinate hierarchical level. Dimensional theories organize emotions according to their relative position along a small number of dimensions. Perhaps the best-known of these is the circumplex model (Russell, 1978). This states that any emotion can be characterized according to its location along two orthogonal dimensions, namely pleasant–unpleasant and arousing–sleepy. For example, tension can be characterized as a combination of high arousal and unpleasantness, whereas serenity can be characterized as a combination of sleepy and pleasantness. Any specific emotion can be conceptualized in terms of a particular quantity of pleasantness and arousal, so, for example, aggressiveness represents a greater amount of arousal than does strength, and elation represents a greater degree of pleasantness than does thankful.

This approach has been used successfully to study emotion in a variety of domains in recent years, including responses to climate change (Leviston, Price, & Bishop, 2014), age differences in temporal variation in emotional state (English & Carstensen, 2014), affective social behavior (Carney & Colvin, 2010), facial expression of emotion (Tseng et al., 2014), and use of music in sports-related motivation (Loizou, Karageorghis, & Bishop, 2014). Moreover, Posner et al. (2009) provide functional magnetic resonance imaging data detailing the neurophysiological bases of pleasantness and arousal in emotion.

Of greatest relevance to the present research, North and Hargraves (1997) found that ratings of pleasantness and participants’ subjective assessment of arousal in response to 32 pieces of music could predict ratings of those same pieces in terms of eight different emotional responses: The results were consistent with the circumplex approach, such that pieces that were liked and arousing were also regarded as exciting, pieces that were disliked and not arousing were also regarded as boring, pieces that were liked and not arousing were regarded as relaxing, and pieces that were disliked and arousing were regarded as aggressive. Subsequent research on emotion in music has produced similar findings. Kreutz, Ott, Teichmann, Osawa, and Vaitl (2008) found that pleasantness and activation ratings of music were related to the specific emotions it elicited; Ritossa and Rickard (2004, see also Madsen, 1998) showed that the emotions expressed by pieces of music could be predicted by a combination of subjective reports of evoked arousal and pleasantness (and also familiarity); and Schubert (2004) identified a link between arousal evoked by music (particularly via loudness and tempo) and emotional responses.

Similarly, other studies show that physiological states indicative of greater physiological arousal are associated with more powerful emotional responses to music (such as experiencing shivers down the spine), just as the circumplex predicts (see reviews by Bartlett, 1996; Scherer & Zentner, 2001): Both Khalfa, Peretz, Blondin, and Manon (2002) and Rickard (2004, see also McFarland, 1985) found that emotionally powerful music gave rise to greater increases in skin conductance than did less emotionally powerful music; Dibben (2004) found that participants who had just exercised reported more intense emotional experiences of music than did participants who had relaxed; and Nyklicek, Thayer, and van Doornen (1997) were able to identify reliable cardio-respiratory responses to different musically induced emotions that were “related to the arousal dimension of self-reported emotions” (p. 304). We should note also, however, that there are instances of contrary findings: For example, Panksepp and Bekkedal’s (1997) electroencephalogram measurements of cortical arousal differed little in response to happy and sad music.

However, Kreutz et al. (2008) and several others have noted that the great majority of research to date has employed lab-based (usually undergraduate) participants listening to relatively short excerpts drawn from small samples of music, which have often been composed or performed specifically for the research. Although there has been some research in music information retrieval that has begun to consider emotion—for example, by overtly considering its role in recommendation systems (e.g., Eerola, Lartillot, & Toiviainen, 2009; Qin, Zheng, Tian, & Zheng, 2014; Scirea, Nelson, & Togelius, 2015) and by specifically considering mood tags (e.g., Laurier, Sordo, Serra, & Herrera, 2009; Saari & Eerola, 2013; Saari et al., 2013). This work has not considered emotion at the population level; and there are similarly exemplars of other research that have used models of emotion that are arguably less widely employed than the circumplex such as categorical models (e.g., using Hevner’s (1936) adjective circle) and domain-specific models (e.g., the Geneva Emotional Music Scales measure)—see Zentner and Eerola (2010) and Zentner, Grandjean, and Scherer et al. (2008). Given the scale of interest in the circumplex approach as a means of explaining emotion in music, and the apparently supportive results among more limited samples of music and participants, there is a clear need to determine whether it can be corroborated in population-wide data that arguably reflects the totality of listening experience. Therefore, in order to carry out such a test, the present research employed a database containing all those pieces that had appeared on one of the UK sales music charts at any point: They represent a complete commercial musical culture.

The literature suggests two hypotheses concerning the relationships between the mood of music and its energy and tempo (representing the arousal-sleepy component of the circumplex), and its popularity (since this is arguably a population-wide proxy for the pleasantness dimension of the circumplex, although we return to this point shortly). Hypothesis 1 was that we might expect that energy and beats per minute (BPM) would both be associated positively with the pieces expressing the emotions regarded by the circumplex approach as representing high levels of arousal, and negatively with those emotions regarded by the circumplex as toward the sleepy end of the dimension. We were more confident of results satisfying this hypothesis in the case of energy than in the case of BPM, as the former represents a more holistic assessment of the arousal intrinsic to a piece than does BPM (since tempo is only one of several possible factors that contributes to the activity of a piece; Berlyne, 1971).

Hypothesis 2 was that we might expect that hit popularity would be associated positively with the pieces expressing emotions that are positively valenced. We have less confidence in this second hypothesis, however, as there are grounds to suspect that a measure of sales and popularity may not represent a direct test of the pleasantness dimension of the circumplex, and we return to this point in the Discussion section. Nonetheless, data on sales and popularity allow us to also test related questions.

In particular, the research was also able to assess two related subsidiary issues on an exploratory basis, namely whether certain musical genres are more likely to evoke certain emotions rather than others. First, it allows us to test simply whether music that evokes certain moods enjoys greater popularity than does music that evokes other moods. Second, there is a long tradition within music psychology and musicology of attempting to identify certain emotional connotations as a reliable outcome of certain structural musical properties. Perhaps the best-known of these is still Cooke’s (1959; see also Kaminska & Woolf, 2000) theory, which claims that certain melodic patterns have a directly communicative, almost linguistic, property in reliably communicating certain emotions, such that, for example, descending passages to the tonic are analogous to peace or rest, whereas passages moving away from the tonic are analogous to outgoing emotions. Indeed, Bruner (1990; see also Gabrielsson & Juslin, 1996; Gabrielsson & Lindström, 2001; Juslin, 2000, 2005; Juslin & Laukka, 2000, 2003) reviewed numerous studies from the fields of psychology, musicology, and marketing and summarized the various possible iconic meanings that different musical structures may have in terms of time-, pitch-, and texture-related factors. Similarly, Straehley and Loebach (2014) found that the emotional connotations of various musical modes could be captured in terms of their valence and intensity, consistent with the circumplex dimensions of pleasantness and arousal, respectively.

As such, we might expect the musical conventions of differing genres to lead to these genres having significantly different emotional connotations. Confirmation of such would have implications for several specific lines of research. North and Hargreaves (2008) reviewed a number of studies within the public health and criminology literature on how certain musical genres, particularly rap and heavy metal (but also blues, country, and opera—see Stack, 2000, 2002; Stack & Gundlach, 1992), are often associated with negatively valenced emotional responses, and these in turn have been claimed to be the cause of elevated mental health problems and juvenile offending among these individuals. Similarly, research on music therapy has identified significant effects (and notable effect sizes) of musically induced emotion on a range of health-related outcomes, such as the experience of pain (see review by Standley, 1995). Consumer research has shown that using music to induce certain moods among customers can influence their purchasing (e.g., North, Shilcock, & Hargreaves, 2003); and research on everyday music listening has identified that one implication of the digitization and portability of music is that listeners place great value on their ability to control the music they experience and seek to use certain genres to evoke desired emotional responses that are useful in the given context of music listening (Krause, North, & Hewitt, 2014). A more wide-ranging understanding of the relationship between genre and mood, based on the large dataset employed here, could inform all these fields.

Method

Dataset

The research employed an adapted version of a master dataset used extensively within the music industry, with the adaptation created in partnership with a private sector organization. The master database contains information on over 38 million pieces of recorded music, which in effect represents all music recordings ever released on a commercial basis in Europe, North America, and Australasia since the beginning of the 20th century (including recordings of pieces composed in earlier centuries). The master database is compiled by a company, which aggregates information globally from over 400,000 record labels. The master database represents the canonical music catalogue used by radio stations, recording companies, and other media in music programming and other similar activities. On entry into the master dataset, the company concerned classifies each piece into one of 23 genres (namely, alternative or indie, blues, cast recordings or cabaret, children’s, Christian or gospel, classical or opera, comedy or spoken word, country, electronica or dance, folk, instrumental, jazz, Latin, new age, pop, rap or hip hop, reggae or ska, rock, seasonal, soul or R&B, soundtracks, vocal, and world) on the basis of the recording artist in question: The initial classification of an artist incorporates information provided by the recording company in question. Note that tracks classified as “comedy/spoken word” were deleted from the present dataset because the great majority did not contain any music, and any music they contain is clearly not the focus of the remainder. Pieces were also deleted for minority genres, for which there were fewer than 100 exemplars that also had popularity data. Created on March 30, 2015, the subset of this master dataset used in the present research contained 143,353 pieces of music, which were selected as those for which data also existed concerning sales in the United Kingdom, such that the pieces employed were all and only those that had enjoyed any commercial success whatsoever in that country: They represent a complete commercial musical culture.

Energy

The energy value for each piece was calculated via an algorithmic process that produced a score for each in turn based on its specific features: This approach is preferable to assigning scores to individual tracks on the basis of metadata, such as genre classification, as it directly addresses the characteristics of the piece in question. The first step was establishing a set of training tracks, consisting of 100 exemplar calm and 100 exemplar energetic pieces, which were selected by a team comprising two students who were heavy music consumers, a musicologist, and an audio engineer working collaboratively. This set of training tracks was used in order to train an artificial intelligence (AI) process (detailed in U.S. Patent No. 20080021851, 2008; U.S. Patent No. 20100250471, 2010) about the sonic differences between energetic and calm tracks using mathematical vectors based on the combinations of 11 sound properties (e.g., tempo, beat, pitch, and rhythm). Via this AI process, the computer compared each individual exemplar track against the remaining 99 using an algorithm: If in the 10 most acoustically similar tracks (again defined according to 11 computer-analyzed sound properties such as tempo, beat, pitch, and rhythm) there was a majority from the same proposed class as the seed track (i.e., calm vs. energetic), then the target piece was regarded as having been classified appropriately. The initial batch of tracks yielded a successful classification rate of 92%, and the 18 incorrectly classified tracks were then replaced by others in subsequent iterations of the same process until all 200 of the seed tracks could be regarded as classified appropriately by this process. The trained AI process (detailed in U.S. Patent No. 20080021851, 2008; U.S. Patent No. 20100250471, 2010), referred to as an energy classifier, was then used to process every track in the database and assign an energy value to each on the basis of the degree of similarity between its own values on the 11 sound properties and the values of the training tracks. A similarity engine combined scores on 69 differing combinations of the 11 sound properties to determine the degree of similarity between a given piece and the other pieces in the database: This was accomplished by examining the degree of similarity on the values for each of the 69 combinations for each track in turn relative to the remainder of the tracks in the database. Each track was then assigned an energy value based on the similarity values so that the greater the similarity between two tracks the greater the similarity in their energy scores: High values indicate an energetic track while low values indicate a calm track. The research team also carried out a nonstatistical informal human listening test of 1,000 tracks from the entire database, selected via a quasi-random process, which involved checking the face validity of relatively low, moderate, and high energy values produced by the AI system.

Beats per minute

Initially, we tested five different algorithmic measures of BPM for each of the genres employed in the present research. These candidate algorithms were based on the industry-standard open source C++ library developed by the Music Technology Group of Pompeu Fabra University (http://essentia.upf.edu). The outputs of each algorithm were then compared against human ratings of a subsample of tracks from each of the genres. The two algorithms that produced outputs with the highest correlation with the human ratings were then combined and subsequently employed in the present research. The BPM value for each piece was determined via computerized measurements that were taken for each successive 30-s segment of each track to allow for rallentando and other forms of tempo variation within the track. The tempo values for each segment were subsequently averaged to provide a single BPM value per piece. Once values had been calculated for each track, the same informal human listening test as described under the Energy subheading indicated that the outputs of this process have good face validity, as they provide a good overall assessment of tempo; and separate unpublished tests of the accuracy of the process (vs. manual measurements of tempo) carried out prior to commencement of the current research also suggest that this approach performs well.

Hit popularity

Each piece was assigned a hit popularity score that utilized data from the U.K. charts at both regional and national level. The measures incorporated data from general charts as well as genre-specific and regional charts. Each chart was assigned a weighting based on the size of the region covered (e.g., a national chart was weighted heavier than a regional chart, with the extent of the difference depending on the size of the region in question); whether the chart addressed singles or albums (with singles charts weighted heavier albums charts, as they are a more direct reflection of the popularity of the specific track in question); and whether the chart was general versus genre- or region-specific (with the extent of the difference in weighting of specific genre charts depending on the popularity of the genre and size of the region in question). For example, the U.K. singles chart was assigned a weighting of 1; the corresponding albums charts were assigned a weighting of .500 (i.e., 1/2); the U.K. classical specialist albums chart was assigned a weighting of .167 (i.e., 1/6); the U.K. Asian singles chart was assigned a weighting of .143 (i.e., 1/7); and the Scottish albums chart was assigned a weighting of .125 (i.e., 1/8). For each track per chart, the popularity score was calculated as 1 divided by (peak chart position multiplied by chart weighting), so that higher scores indicate greater popularity.

Mood scores

Each track was assigned values for each of six moods, represented by numbered adjective clusters, namely Mood 1 = clean, simple, and relaxing; Mood 2 = happy, hopeful, and ambition; Mood 3 = passion, romance, and power; Mood 4 = mystery, luxury, and comfort; Mood 5 = energetic, bold, and outgoing; and Mood 6 = calm, peace, and tranquility, respectively. These moods were employed at the discretion of the music industry at the time the initial database was devised and are regarded by the industry as most relevant to radio programming (and similar commercial uses): Nonetheless, they possess good face validity as typical responses to music and map well onto previous research on the circumplex, so that Moods 1, 4, and 6 are located at the lower end of the arousal dimension whereas Moods 2, 3, and 5 are located at the higher end of this dimension. Unfortunately, however, these moods do not reflect the negative end of the pleasantness dimension.

The mood scores were based on seed ratings of 300 pieces thought to represent a good range of all the moods concerned. Again, to begin the process of processing the scores, six musicians and sound engineers participated in an informal exercise that provided ratings of how the music made them feel in order to create a training set of tracks for the AI training. The development of the mood scores involved a three-step machine learning process, similar to that for the Energy score (U.S. Patent No. 20080021851, 2008; U.S. Patent No. 20100250471, 2010). First, each piece was analyzed according to audio descriptors based on melody, harmony, tempo, pitch, octave, beat, rhythm, noise, brilliance, and chord progression. Second, as per the energy score, a similarity engine combined scores on 69 differing combinations of the audio descriptors to determine the extent to which each track was similar to the others in the database. Third, each of the six mood scores for each piece was then determined on the basis of the mood scores assigned to similar tracks and the degree of similarity between those and the target piece on the 69 combinations of the audio descriptors. This allowed the computer to allocate percentage scores to each track that represented the extent to which it fitted each of the six moods, so that the higher the mood score in question so the more that the piece represented that mood (since it shared sonic characteristics with other pieces that represented the same mood). The same informal human listening test as described under the Energy subheading indicated that the outputs of this process have good face validity.

Results

Energy, BPM, Hit Popularity, and Mood

A series of General Linear Mixed Model (GLMM) analyses addressed the first and second hypotheses, namely whether energy, BPM, and hit popularity could predict scores on each of the six moods (α < .001, to allow for the multiple analyses performed). Energy, BPM, and hit popularity served as predictor variables in six separate GLMM analyses concerning each of the mood scores in turn, respectively (see Table 1(a) to (f)). The effect sizes indicate that energy explained a much greater portion of the variance (ranging between 5% and 28%) than did BPM or hit popularity. This set of six analyses was then repeated for each genre separately (α < .001; see Table 1(a) to (f)). These analyses again indicated that energy predicted a greater portion of the variance in the mood scores than did BPM or hit popularity.

Table 1a.

GLMM Analysis Predicting Mood 1 Scores (Clean, Simple, and Relaxing).

Analysis variables	F	DF	DF_error	p	β	t	95% CI	η²
Overall dataset (N = 143,353)
Corrected model	4214.53	3	143349	<.001
Energy	12544.01	1	143349	<.001	−0.04	−112.00	[−0.04, −0.04]	0.080
BPM	28.06	1	143349	<.001	0.00	5.30	[0.00, 0.00]	0.000
Hit popularity	14.16	1	143349	<.001	0.16	3.76	[0.08, 0.25]	0.000
Alternative or indie (N = 806)
Corrected model	60.54	3	802	<.001
Energy	153.45	1	802	<.001	−0.03	−12.39	[−0.04, −0.03]	0.161
BPM	9.83	1	802	.002	−0.01	−3.14	[−0.17, 0.00]	0.012
Hit popularity	0.71	1	802	.399	0.97	0.84	[−1.29, 3.24]	0.001
Christian or gospel (N = 222)
Corrected model	1.31	3	218	.273
Energy	2.83	1	218	.094	−0.04	−1.68	[−0.08, 0.01]	0.013
BPM	0.54	1	218	.465	0.01	0.73	[−0.02, 0.03]	0.002
Hit popularity	0.18	1	218	.673	−1.84	−0.42	[−10.43, 6.75]	0.001
Classical (N = 4,745)
Corrected model	277.49	3	4741	<.001
Energy	816.7	1	4741	<.001	−0.26	−28.58	[−0.28, −0.24]	0.147
BPM	2.49	1	4741	.114	0.01	1.58	[0.00, 0.01]	0.001
Hit popularity	10.2	1	4741	.001	6.74	3.19	[2.60, 10.87]	0.002
Country (N = 2,552)
Corrected model	19.62	3	2548	<.001
Energy	53.89	1	2548	<.001	−0.03	−7.34	[−0.03, −0.02]	0.021
BPM	2.47	1	2548	.116	0.00	−1.57	[−0.01, 0.00]	0.001
Hit popularity	0.49	1	2548	.483	0.58	0.70	[−1.04, 2.19]	0.000
Electronica or dance (N = 16,086)
Corrected model	84.74	3	16082	<.001
Energy	8.86	1	16082	.003	0.00	2.98	[0.00, 0.00]	0.001
BPM	215.87	1	16082	<.001	0.01	14.69	[0.01, 0.01]	0.013
Hit popularity	10.79	1	16082	.001	0.32	3.29	[0.13, 0.51]	0.001
Folk (N = 992)
Corrected model	43.72	3	988	<.001
Energy	131.09	1	988	<.001	−0.08	−11.45	[−0.09, −0.06]	0.117
BPM	0.16	1	988	.692	0.00	0.40	[−0.01, 0.01]	0.000
Hit popularity	0.3	1	988	.583	−0.36	−0.55	[−1.65, 0.93]	0.000
Jazz (N = 4,300)
Corrected model	67.05	3	4296	<.001
Energy	168.95	1	4296	<.001	−0.10	−13.00	[−0.12, −0.09]	0.038
BPM	11.56	1	4296	.001	−0.01	−3.40	[−0.02, −0.01]	0.003
Hit popularity	0.11	1	4296	.746	−0.58	−0.32	[−4.12, 2.96]	0.000
Latin (N = 633)
Corrected model	13.13	3	629	<.001
Energy	36.36	1	629	<.001	−0.02	−6.03	[−0.02, −0.01]	0.055
BPM	3.09	1	629	.079	−0.01	−1.76	[−0.01, 0.00]	0.005
Hit popularity	0.05	1	629	.829	0.10	0.22	[−0.77, 0.96]	0.000
Pop (N = 58,250)
Corrected model	806.24	3	58246	<.001
Energy	2095.14	1	58246	<.001	−0.02	−45.77	[−0.02, −0.02]	0.035
BPM	176.53	1	58246	<.001	−0.01	−13.29	[−0.01, −0.01]	0.003
Hit popularity	24.99	1	58246	<.001	0.17	5.00	[0.10, 0.23]	0.000
Rap or hip hop (N = 8,296)
Corrected model	2.3	3	8292	.075
Energy	2.96	1	8292	.085	0.00	−1.72	[0.00, 0.00]	0.000
BPM	0.3	1	8292	.584	0.00	−0.55	[0.00, 0.00]	0.000
Hit popularity	3.65	1	8292	.056	0.08	1.91	[0.00, 0.17]	0.000
Reggae or ska (N = 215)
Corrected model	1.94	3	211	.124
Energy	2.89	1	211	.091	0.01	1.70	[0.00, 0.02]	0.014
BPM	0.05	1	211	.817	0.00	−0.23	[−0.01, 0.01]	0.000
Hit popularity	2.49	1	211	.116	5.55	3.51	[−1.38, 12.47]	0.055
Rock (N = 44,307)
Corrected model	323.55	3	44303	<.001
Energy	730.25	1	44303	<.001	−0.01	−27.02	[−0.01, −0.01]	0.016
BPM	137.05	1	44303	<.001	0.00	−11.71	[0.00, 0.00]	0.003
Hit popularity	45.49	1	44303	<.001	0.36	6.74	[0.26, 0.47]	0.001
Soul or R&B (N = 869)
Corrected model	28.19	3	865	<.001
Energy	64.25	1	865	<.001	−0.02	−8.02	[−0.02, −0.01]	0.069
BPM	2.99	1	865	.084	0.00	−1.73	[−0.01, 0.00]	0.003
Hit popularity	12.05	1	865	.001	2.10	3.47	[0.91, 3.28]	0.014
Soundtracks (N = 406)
Corrected model	8.6	3	402	<.001
Energy	14.13	1	402	<.001	−0.13	−3.76	[−0.20, −0.06]	0.034
BPM	0.47	1	402	.493	0.01	0.69	[−0.01, 0.03]	0.001
Hit popularity	7.72	1	402	.006	26.62	2.78	[7.79, 45.46]	0.019
World (N = 542)
Corrected model	21.49	3	538	<.001
Energy	61.46	1	538	<.001	−0.06	−7.84	[−0.07, −0.04]	0.103
BPM	2.09	1	538	.149	0.01	1.45	[0.00, 0.03]	0.004
Hit popularity	0.76	1	538	.385	2.81	0.87	[−3.54, 9.16]	0.001

DF = degrees of freedom; CI = confidence interval; BPM = beats per minute.

Mood by Genre

A second set of six GLMM analyses (α < .001, to allow for the multiple analyses) considered variations between genres on each of the six mood scores, respectively. All six analyses were significant, with the associated deviation contrasts demonstrating the scores for each genre relative to the overall mean score per mood. These means and details pertaining to the deviation contrasts for each genre are presented in Table 2(a) to (f).

Table 1b.

GLMM Analysis Predicting Mood 2 Scores (Happy, Hopeful, and Ambition).

Analysis variables	F	DF	DF_error	p	β	t	95% CI	η²
Overall dataset (N = 143,353)
Corrected model	3855.90	3	143349	<.001
Energy	10962.94	1	143349	<.001	−0.04	−104.70	[−0.04, −0.04]	0.071
BPM	54.06	1	143349	<.001	0.00	−7.35	[0.00, 0.00]	0.000
Hit popularity	94.39	1	143349	<.001	−0.43	−9.72	[−0.52, −0.34]	0.001
Alternative or indie (N = 806)
Corrected model	3.39	3	802	<.001
Energy	4.64	1	802	.032	0.01	2.15	[0.00, 0.01]	0.006
BPM	1.02	1	802	.314	0.00	−1.01	[−0.01, 0.00]	0.001
Hit popularity	4.83	1	802	.028	−2.68	−2.20	[−5.06, −0.29]	0.006
Christian or gospel (N = 222)
Corrected model	0.78	3	218	.504
Energy	1.90	1	218	.169	0.03	1.38	[−0.01, 0.06]	0.009
BPM	0.01	1	218	.924	0.00	0.10	[−0.02, 0.02]	0.000
Hit popularity	0.84	1	218	.360	−3.16	−0.92	[−9.94, 3.62]	0.004
Classical or opera (N = 4,745)
Corrected model	68.25	3	4741	<.001
Energy	18.72	1	4741	<.001	0.05	13.59	[0.04, 0.05]	0.038
BPM	0.76	1	4741	.384	0.00	0.87	[0.00, 0.00]	0.000
Hit popularity	15.50	1	4741	<.001	−3.09	−3.94	[−4.64, −1.55]	0.003
Country (N = 2,552)
Corrected model	22.02	3	2548	<.001
Energy	47.02	1	2548	<.001	0.04	6.86	[0.03, 0.05]	0.018
BPM	13.03	1	2548	<.001	0.01	3.61	[0.01, 0.02]	0.005
Hit popularity	1.55	1	2548	.214	1.44	1.24	[−0.83, 3.70]	0.001
Electronica or dance (N = 16,086)
Corrected model	123.68	3	16082	<.001
Energy	94.39	1	16082	<.001	−0.01	−9.72	[−0.01, −0.01]	0.006
BPM	212.98	1	16082	<.001	−0.02	−14.59	[−0.02, −0.01]	0.013
Hit popularity	12.33	1	16082	<.001	−0.72	−3.51	[−1.13, −0.32]	0.001
Folk (N = 992)
Corrected model	73.38	3	988	<.001
Energy	217.79	1	988	<.001	0.12	14.76	[0.11, 0.14]	0.181
BPM	0.33	1	988	.568	0.00	0.57	[−0.01, 0.01]	0.000
Hit popularity	0.00	1	988	.949	−0.05	−0.06	[−1.66, 1.55]	0.000
Jazz (N = 4,300)
Corrected model	374.44	3	4296	<.001
Energy	985.13	1	4296	<.001	0.69	31.39	[0.16, 0.18]	0.187
BPM	25.51	1	4296	<.001	0.01	5.05	[0.01, 0.02]	0.006
Hit popularity	9.73	1	4296	.002	−3.92	−3.12	[−6.38, −1.46]	0.002
Latin (N = 633)
Corrected model	10.85	3	629	<.001
Energy	22.03	1	629	<.001	−0.03	−4.69	[−0.04, −0.01]	0.034
BPM	8.59	1	629	.004	0.02	2.93	[0.01, 0.03]	0.013
Hit popularity	0.95	1	629	.330	−0.81	−0.98	[−2.43, 0.82]	0.002
Pop (N = 58,250)
Corrected model	407.83	3	58246	<.001
Energy	953.41	1	58246	<.001	−0.02	−30.88	[−0.02, −0.02]	0.016
BPM	253.07	1	58246	<.001	0.01	15.91	[0.01, 0.01]	0.004
Hit popularity	82.13	1	58246	<.001	−0.49	−9.06	[−0.59, −0.38]	0.001
Rap or hip hop (N = 8,296)
Corrected model	12.92	3	8292	<.001
Energy	11.62	1	8292	.001	−0.01	−3.41	[−0.01, 0.00]	0.001
BPM	17.73	1	8292	<.001	−0.01	−4.21	[−0.01, 0.00]	0.002
Hit popularity	9.00	1	8292	.003	−0.35	−3.00	[−0.58, −0.12]	0.001
Reggae or ska (N = 215)
Corrected model	3.76	3	211	.012
Energy	5.91	1	211	.016	−0.06	−2.43	[−0.10, −0.01]	0.027
BPM	1.20	1	211	.275	0.02	1.09	[−0.01, 0.05]	0.006
Hit popularity	3.75	1	211	.054	−26.00	−1.94	[−52.46, 0.47]	0.017
Rock (N = 44,307)
Corrected model	3028.43	3	44303	<.001
Energy	8933.37	1	44303	<.001	−0.05	−94.52	[−0.05, −0.05]	0.168
BPM	0.03	1	44303	.867	0.00	−0.17	[0.00, 0.00]	0.000
Hit popularity	37.99	1	44303	<.001	−0.63	−6.16	[−0.84, −0.43]	0.001
Soul or R&B (N = 869)
Corrected model	11.38	3	865	<.001
Energy	14.36	1	865	<.001	0.03	3.79	[0.01, 0.04]	0.016
BPM	7.09	1	865	.008	0.01	2.66	[0.00, 0.02]	0.008
Hit popularity	9.44	1	865	.002	−5.73	−3.07	[−9.39, −2.07]	0.011
Soundtracks (N = 406)
Corrected model	11.50	3	402	<.001
Energy	22.95	1	402	<.001	0.09	4.79	[0.05, 0.12]	0.054
BPM	1.28	1	402	.259	−0.01	−1.13	[−0.02, 0.01]	0.003
Hit popularity	5.81	1	402	.016	−12.24	−2.41	[−22.23, −2.26]	0.014
World (N = 542)
Corrected model	7.11	3	538	<.001
Energy	19.37	1	538	<.001	0.03	4.40	[0.02, 0.05]	0.035
BPM	0.01	1	538	.923	0.00	−0.10	[−0.01, 0.01]	0.000
Hit popularity	2.26	1	538	.134	−4.63	−1.50	[−10.69, 1.43]	0.004

DF = degrees of freedom; CI = confidence interval; BPM = beats per minute.

Table 1c.

GLMM Analysis Predicting Mood 3 Scores (Passion, Romance, and Power).

Analysis variables	F	DF	DF_error	p	β	t	95% CI	η²
Overall dataset (N = 143,353)
Corrected model	18440.83	3	143349	<.001
Energy	52437.41	1	143349	<.001	0.15	228.99	[0.15, 0.15]	0.268
BPM	502.58	1	143349	<.001	0.02	22.42	[0.02, 0.02]	0.003
Hit popularity	25.44	1	143349	<.001	−0.42	−5.04	[−0.58, −0.26]	0.000
Alternative or indie (N = 806)
Corrected model	272.50	3	802	<.001
Energy	720.15	1	802	<.001	0.17	26.84	[0.16, 0.18]	0.473
BPM	29.58	1	802	<.001	0.05	5.44	[0.03, 0.06]	0.036
Hit popularity	0.30	1	802	.582	−1.59	−0.55	[−7.25, 4.07]	0.000
Christian or gospel (N = 222)
Corrected model	15.17	3	218	<.001
Energy	29.41	1	218	<.001	0.14	5.42	[0.09, 0.20]	0.119
BPM	0.45	1	218	.502	0.01	0.67	[−0.02, 0.04]	0.002
Hit popularity	7.47	1	218	.007	13.71	2.73	[3.82, 23.59]	0.033
Classical or opera (N = 4,745)
Corrected model	351.79	3	4741	<.001
Energy	1047.66	1	4741	<.001	0.18	32.37	[0.17, 0.19]	0.181
BPM	0.22	1	4741	.638	0.00	−0.47	[−0.01, 0.00]	0.000
Hit popularity	3.82	1	4741	.051	−2.56	−1.95	[−5.14, 0.01]	0.001
Country (N = 2,552)
Corrected model	173.17	3	2548	<.001
Energy	490.56	1	2548	<.001	0.15	22.15	[0.13, 0.16]	0.161
BPM	12.39	1	2548	<.001	0.01	3.52	[0.01, 0.02]	0.005
Hit popularity	2.33	1	2548	.127	−2.29	−1.53	[−5.23, 0.65]	0.001
Electronica or dance (N = 16,086)
Corrected model	1675.83	3	16082	<.001
Energy	4883.12	1	16082	<.001	0.09	69.88	[0.09, 0.10]	0.233
BPM	1.81	1	16082	.178	0.00	−1.35	[0.00, 0.00]	0.000
Hit popularity	36.73	1	16082	<.001	1.52	6.06	[1.03, 2.01]	0.002
Folk (N = 992)
Corrected model	66.15	3	988	<.001
Energy	193.41	1	988	<.001	0.14	13.91	[0.12, 0.16]	0.164
BPM	0.89	1	988	.346	0.01	0.94	[−0.01, 0.02]	0.001
Hit popularity	0.76	1	988	.385	−0.84	−0.87	[−2.75, 1.06]	0.001
Jazz (N = 4,300)
Corrected model	293.05	3	4296	<.001
Energy	799.98	1	4296	<.001	0.14	28.28	[0.13, 0.15]	0.157
BPM	15.96	1	4296	<.001	0.01	4.00	[0.00, 0.01]	0.004
Hit popularity	0.72	1	4296	.396	0.98	0.85	[−1.29, 3.25]	0.000
Latin (N = 633)
Corrected model	43.84	3	629	<.001
Energy	125.33	1	629	<.001	0.07	11.20	[0.06, 0.09]	0.166
BPM	0.32	1	629	0.572	0.00	−0.57	[−0.02, 0.01]	0.001
Hit popularity	2.65	1	629	0.104	1.63	1.63	[−0.34, 3.60]	0.004
Pop (N = 58,250)
Corrected model	3777.21	3	58246	<.001
Energy	10973.65	1	58246	<.001	0.10	104.76	[0.09, 0.10]	0.159
BPM	80.10	1	58246	<.001	0.01	8.95	[0.01, 0.01]	0.001
Hit popularity	0.00	1	58246	.979	0.00	−0.03	[−0.15, 0.15]	0.000
Rap or hip hop (N = 8,296)
Corrected model	1067.68	3	8292	<.001
Energy	3188.49	1	8292	<.001	0.07	56.47	[0.07, 0.08]	0.278
BPM	2.08	1	8292	.149	0.00	1.44	[0.00, 0.00]	0.000
Hit popularity	8.70	1	8292	.003	0.28	2.95	[0.10, 0.47]	0.001
Reggae or ska (N = 215)
Corrected model	17.09	3	211	<.001
Energy	49.66	1	211	<.001	0.06	7.05	[0.05, 0.08]	0.191
BPM	0.05	1	211	.829	0.00	0.22	[−0.01, 0.01]	0.000
Hit popularity	0.21	1	211	.651	2.37	0.45	[−7.95, 12.68]	0.001
Rock (N = 44,307)
Corrected model	5700.66	3	44303	<.001
Energy	16293.03	1	44303	<.001	0.14	127.64	[0.14, 0.14]	0.269
BPM	282.19	1	44303	<.001	0.02	16.80	[0.02, 0.02]	0.006
Hit popularity	20.86	1	44303	<.001	−0.97	−4.57	[−1.39, −0.56]	0.000
Soul or R&B (N = 869)
Corrected model	36.61	3	865	<.001
Energy	85.35	1	865	<.001	0.10	9.24	[0.08, 0.12]	0.090
BPM	7.78	1	865	.005	0.02	2.79	[0.01, 0.04]	0.009
Hit popularity	7.86	1	865	.005	−7.85	−2.80	[−13.3, −2.35]	0.009
Soundtracks (N = 406)
Corrected model	18.23	3	402	<.001
Energy	49.85	1	402	<.001	0.24	7.06	[0.17, 0.30]	0.110
BPM	0.27	1	402	.607	−0.01	−0.51	[−0.03, 0.01]	0.001
Hit popularity	0.98	1	402	.323	−9.20	−0.99	[−27.45, 9.06]	0.002
World (N = 542)
Corrected model	22.64	3	538	<.001
Energy	64.46	1	538	<.001	0.07	8.03	[0.06, 0.09]	0.107
BPM	3.97	1	538	.047	0.02	1.99	[0.00, 0.04]	0.007
Hit popularity	0.07	1	538	.797	1.02	0.26	[−6.78, 8.83]	0.000

DF = degrees of freedom; CI = confidence interval; BPM = beats per minute.

Table 1d.

GLMM Analysis Predicting Mood 4 Scores (Mystery, Luxury, and Comfort).

Analysis variables	F	DF	DF_error	p	β	t	95% CI	η²
Overall dataset (N = 143,353)
Corrected model	5496.7	3	143349	<.001
Energy	14731.71	1	143349	<.001	−0.04	−121.37	[−0.04, −0.04]	0.093
BPM	621.82	1	143349	<.001	−0.01	−24.94	[−0.10, −0.01]	0.004
Hit popularity	50.77	1	143349	<.001	0.30	7.13	[0.22, 0.38]	0.000
Alternative or indie (N = 806)
Corrected model	18.84	3	802	<.001
Energy	22.58	1	802	<.001	−0.01	−4.75	[−0.02, −0.01]	0.027
BPM	18.11	1	802	<.001	−0.02	−4.26	[−0.02, −0.01]	0.022
Hit popularity	6.63	1	802	.010	3.39	2.58	[0.81, 5.98]	0.008
Christian or gospel (N = 222)
Corrected model	2.05	3	218	.108
Energy	0.37	1	218	.543	−0.01	−0.61	[−0.06, 0.03]	0.002
BPM	1.57	1	218	.212	−0.02	−1.25	[−0.04, 0.01]	0.007
Hit popularity	3.61	1	218	.059	−8.41	−1.90	[−17.12, 0.31]	0.016
Classical (N = 4,745)
Corrected model	20.37	3	4741	<.001
Energy	47.79	1	4741	<.001	−0.04	−6.91	[−0.05, −0.03]	0.010
BPM	1.17	1	4741	.280	0.00	−1.08	[−0.01, 0.00]	0.000
Hit popularity	13.25	1	4741	<.001	−5.03	−3.64	[−7.74, −2.32]	0.003
Country (N = 2,552)
Corrected model	22.09	3	2548	<.001
Energy	49.72	1	2548	<.001	−0.04	−7.05	[−0.05, −0.03]	0.019
BPM	10.62	1	2548	.001	−0.01	−3.26	[−0.02, −0.01]	0.004
Hit popularity	2.2	1	2548	.138	1.96	1.48	[−0.63, 4.55]	0.001
Electronica or dance (N = 16,086)
Corrected model	1019.85	3	16082	<.001
Energy	2549.68	1	16082	<.001	−0.04	−50.49	[−0.04, −0.04]	0.137
BPM	221.66	1	16082	<.001	−0.01	−14.89	[−0.01, −0.01]	0.014
Hit popularity	0.03	1	16082	.861	0.03	0.18	[−0.28, 0.33]	0.000
Folk (N = 992)
Corrected model	20.06	3	988	<.001
Energy	59.55	1	988	<.001	−0.07	−7.72	[−0.09, −0.05]	0.057
BPM	0	1	988	.995	0.00	0.01	[−0.01, 0.01]	0.000
Hit popularity	0.13	1	988	.715	0.32	0.37	[−1.41, 2.06]	0.000
Jazz (N = 4,300)
Corrected model	171.62	3	4296	<.001
Energy	489.87	1	4296	<.001	−0.19	−22.13	[−0.21, −0.17]	0.102
BPM	1.54	1	4296	.214	0.00	−1.24	[−0.01, 0.00]	0.000
Hit popularity	1.43	1	4296	.232	−2.38	−1.20	[−6.27, 1.52]	0.000
Latin (N = 633)
Corrected model	21.3	3	629	<.001
Energy	62.48	1	629	<.001	−0.04	−7.90	[−0.05, −0.03]	0.090
BPM	1.28	1	629	.259	0.01	1.13	[−0.01, 0.02]	0.002
Hit popularity	0.1	1	629	.757	0.24	0.31	[−1.30, 1.79]	0.000
Pop (N = 58,250)
Corrected model	958.37	3	58246	<.001
Energy	2613.39	1	58246	<.001	−0.03	−51.12	[−0.03, −0.03]	0.043
BPM	122.61	1	58246	<.001	−0.01	−11.07	[−0.01, −0.01]	0.002
Hit popularity	20.72	1	58246	<.001	0.23	4.55	[0.13, 0.33]	0.000
Rap or hip hop (N = 8,296)
Corrected model	375.62	3	8292	<.001
Energy	1087.56	1	8292	<.001	−0.04	−32.98	[−0.04,−0.03]	0.116
BPM	34.69	1	8292	<.001	−0.01	−5.89	[−0.01, 0.00]	0.004
Hit popularity	0.02	1	8292	.895	0.01	0.13	[−0.15, 0.17]	0.000
Reggae or ska (N = 215)
Corrected model	1.73	3	211	.162
Energy	4.47	1	211	.036	−0.03	−2.11	[−0.06, 0.00]	0.021
BPM	0.43	1	211	.512	−0.01	−0.66	[−0.03, 0.01]	0.002
Hit popularity	0.01	1	211	.919	0.84	0.10	[−15.30, 16.98]	0.000
Rock (N = 44,307)
Corrected model	136.01	3	44303	<.001
Energy	14.7	1	44303	<.001	0.00	3.83	[0.00, 0.00]	0.000
BPM	303.24	1	44303	<.001	−0.01	−17.41	[−0.01, −0.01]	0.007
Hit popularity	91.24	1	44303	<.001	0.77	9.55	[0.61, 0.92]	0.002
Soul or R&B (N = 869)
Corrected model	38.7	3	865	<.001
Energy	107.9	1	865	<.001	−0.09	−10.39	[−0.10, −0.07]	0.111
BPM	1.69	1	865	.194	−0.01	−1.30	[−0.02, 0.00]	0.002
Hit popularity	0.9	1	865	.343	2.02	0.95	[−2.16, 6.21]	0.001
Soundtracks (N = 406)
Corrected model	6.3	3	402	<.001
Energy	18.14	1	402	<.001	−0.08	−4.26	[−0.12, −0.04]	0.043
BPM	0.44	1	402	.508	0.00	−0.66	[−0.02, 0.01]	0.001
Hit popularity	1.83	1	402	.177	−6.97	−1.35	[−17.11, 3.17]	0.005
World (N = 542)
Corrected model	30.75	3	538	<.001
Energy	88.59	1	538	<.001	−0.09	v9.41	[−0.10, −0.07]	0.141
BPM	0.58	1	538	.445	−0.01	−0.76	[−0.02, 0.01]	0.001
Hit popularity	2.57	1	538	.110	−6.36	−1.60	[−14.15, 1.44]	0.005

DF = degrees of freedom; CI = confidence interval; BPM = beats per minute.

Table 1e.

GLMM Analysis Predicting Mood 5 Scores (Energetic, Bold, and Outgoing).

Analysis variables	F	DF	DF_error	p	β	t	95% CI	η²
Overall dataset (N = 143,353)
Corrected model	2884.50	3	143349	<.001
Energy	8435.54	1	143349	<.001	0.04	91.85	[0.04, 0.05]	0.056
BPM	1.74	1	143349	.187	0.00	−1.32	[0.00, 0.00]	0.000
Hit popularity	131.77	1	143349	<.001	−0.68	−11.48	[−0.80, −0.57]	0.001
Alternative or indie (N = 806)
Corrected model	48.23	3	802	<.001
Energy	113.25	1	802	<.001	0.04	10.64	[0.03, 0.05]	0.124
BPM	7.20	1	802	.007	0.01	2.68	[0.00, 0.02]	0.009
Hit popularity	8.99	1	802	.003	−5.13	−3.00	[−8.48, −1.77]	0.011
Christian or gospel (N = 222)
Corrected model	8.38	3	218	<.001
Energy	24.27	1	218	<.001	0.10	4.93	[0.06, 0.14]	0.100
BPM	0.01	1	218	.918	0.00	−0.10	[−0.02, 0.02]	0.000
Hit popularity	3.65	1	218	.057	−7.40	−1.91	[−15.03, 0.23]	0.016
Classical or opera (N = 4,745)
Corrected model	412.89	3	4741	<.001
Energy	1233.72	1	4741	<.001	0.11	35.12	[0.11, 0.12]	0.206
BPM	1.60	1	4741	.206	0.00	1.26	[0.00, 0.01]	0.000
Hit popularity	0.03	1	4741	.864	0.13	0.17	[−1.34, 1.60]	0.000
Country (N = 2,552)
Corrected model	23.96	3	2548	<.001
Energy	36.26	1	2548	<.001	0.04	6.02	[0.03, 0.05]	0.014
BPM	9.20	1	2548	.002	0.01	3.03	[0.00, 0.02]	0.004
Hit popularity	23.85	1	2548	<.001	−6.76	−4.88	[−9.48, −4.05]	0.009
Electronica or dance (N = 16,086)
Corrected model	1183.01	3	16082	<.001
Energy	3503.77	1	16082	<.001	0.09	59.19	[0.09, 0.10]	0.179
BPM	233.17	1	16082	<.001	−0.02	−15.27	[−0.03, −0.02]	0.014
Hit popularity	6.82	1	16082	.009	−0.78	−2.61	[−1.36, −0.19]	0.000
Folk (N = 992)
Corrected model	129.24	3	988	<.001
Energy	380.91	1	988	<.001	0.19	19.52	[0.17, 0.21]	0.278
BPM	1.19	1	988	.277	0.01	1.09	[−0.01, 0.02]	0.001
Hit popularity	0.43	1	988	.513	−0.62	−0.65	[−2.48, 1.24]	0.000
Jazz (N = 4,300)
Corrected model	452.65	3	4296	<.001
Energy	1245.65	1	4296	<.001	0.21	35.29	[0.20, 0.22]	0.225
BPM	15.86	1	4296	<.001	0.01	3.98	[0.01, 0.02]	0.004
Hit popularity	0.92	1	4296	.337	−1.33	−0.96	[−4.04, 1.38]	0.000
Latin (N = 633)
Corrected model	27.91	3	629	<.001
Energy	81.74	1	629	<.001	0.07	9.04	[0.06, 0.09]	0.115
BPM	0.59	1	629	.443	0.01	0.77	[−0.01, 0.03]	0.001
Hit popularity	3.29	1	629	.070	−2.28	−1.81	[−4.75, 0.19]	0.005
Pop (N = 58,250)
Corrected model	2680.01	3	58246	<.001
Energy	7307.93	1	58246	<.001	0.07	85.49	[0.07, 0.07]	0.111
BPM	315.95	1	58246	<.001	0.02	17.78	[0.02, 0.02]	0.005
Hit popularity	110.60	1	58246	<.001	−0.76	−10.52	[−0.90, −0.62]	0.002
Rap or hip hop (N = 8,296)
Corrected model	412.67	3	8292	<.001
Energy	1234.03	1	8292	<.001	0.07	35.13	[0.06, 0.07]	0.130
BPM	0.84	1	8292	.359	0.00	−0.92	[−0.01, 0.00]	0.000
Hit popularity	4.32	1	8292	.038	−0.30	−2.08	[−0.59, −0.02]	0.001
Reggae or ska (N = 215)
Corrected model	3.29	3	211	.022
Energy	0.13	1	211	.715	0.01	0.37	[−0.03, 0.04]	0.001
BPM	4.49	1	211	.035	0.03	2.12	[0.00, 0.06]	0.021
Hit popularity	4.96	1	211	.027	−24.59	−2.23	[−46.36, −2.82]	0.023
Rock (N = 44,307)
Corrected model	93.33	3	44303	<.001
Energy	187.37	1	44303	<.001	−0.01	−13.69	[−0.01, −0.01]	0.004
BPM	15.27	1	44303	<.001	0.00	3.91	[0.00, 0.00]	0.000
Hit popularity	77.35	1	44303	<.001	−1.29	−8.80	[−1.58, −1.00]	0.002
Soul or R&B (N = 869)
Corrected model	112.47	3	865	<.001
Energy	313.07	1	865	<.001	0.17	17.69	[0.16, 0.19]	0.266
BPM	0.92	1	865	.339	0.01	0.96	[−0.01, 0.02]	0.001
Hit popularity	12.60	1	865	<.001	−8.82	−3.55	[−13.69, −3.94]	0.014
Soundtracks (N = 406)
Corrected model	15.40	3	402	<.001
Energy	36.06	1	402	<.001	0.11	6.01	[−0.09, −5.29]	0.082
BPM	4.35	1	402	.038	−0.01	−2.09	[−0.01, −2.70]	0.011
Hit popularity	1.10	1	402	.296	−5.33	−1.05	[−3.80, −0.79]	0.003
World (N = 542)
Corrected model	36.29	3	538	<.001
Energy	106.81	1	538	<.001	0.09	10.34	[0.07, 0.11]	0.166
BPM	0.45	1	538	.505	0.01	0.67	[−0.01, 0.02]	0.001
Hit popularity	2.62	1	538	.106	−6.08	−1.62	[−13.46, 1.30]	0.005

DF = degrees of freedom; CI = confidence interval; BPM = beats per minute.

Table 1f.

GLMM Analysis Predicting Mood 6 Scores (Calm, Peace, and Tranquility).

Analysis variables	F	DF	DF_error	p	β	t	95% CI	η²
Overall dataset (N = 143,353)
Corrected model	19221.98	3	143349	<.001
Energy	54609.07	1	143349	<.001	−0.08	−233.69	[−0.08, −0.08]	0.276
BPM	539.11	1	143349	<.001	−0.01	−23.22	[−0.01, −0.01]	0.004
Hit popularity	0.24	1	143349	.626	0.02	0.49	[−0.06, 0.10]	0.000
Alternative or indie (N = 806)
Corrected model	207.89	3	802	<.001
Energy	543.71	1	802	<.001	−0.08	−23.32	[−0.09, −0.07]	0.404
BPM	19.03	1	802	<.001	−0.02	−4.36	[−0.03, −0.01]	0.023
Hit popularity	7.94	1	802	.005	4.30	2.82	[1.30, 7.29]	0.010
Christian or gospel (N = 222)
Corrected model	16.62	3	218	<.001
Energy	41.16	1	218	<.001	−0.14	−6.42	[−0.18, −0.09]	0.159
BPM	0.05	1	218	.821	0.00	0.23	[−0.02, 0.03]	0.000
Hit popularity	2.46	1	218	.119	−6.33	−1.57	[−14.28, 1.63]	0.011
Classical or opera (N = 4,745)
Corrected model	41.21	3	4741	<.001
Energy	83.29	1	4741	<.001	−0.04	−9.13	[−0.05, −0.03]	0.017
BPM	28.22	1	4741	<.001	0.00	−5.31	[−0.02, −0.01]	0.006
Hit popularity	12.53	1	4741	<.001	−3.71	−3.54	[−5.76, −1.65]	0.003
Country (N = 2,552)
Corrected model	79.50	3	2548	<.001
Energy	218.84	1	2548	<.001	−0.11	−14.79	[−0.13, −0.10]	0.079
BPM	8.18	1	2548	.004	−0.01	−2.86	[−0.02, 0.00]	0.003
Hit popularity	4.50	1	2548	.034	3.65	2.12	[0.28, 7.01]	0.002
Electronica or dance (N = 16,086)
Corrected model	537.77	3	16082	<.001
Energy	1367.78	1	16082	<.001	−0.03	−36.98	[−0.03, −0.02]	0.078
BPM	100.83	1	16082	<.001	−0.01	−10.04	[−0.01, −0.01]	0.006
Hit popularity	3.67	1	16082	.055	0.25	1.92	[−0.01, 0.50]	0.000
Folk (N = 992)
Corrected model	127.53	3	988	<.001
Energy	379.35	1	988	<.001	−0.21	−19.48	[−0.23, −0.19]	0.277
BPM	0.21	1	988	.646	0.00	−0.46	[−0.02, 0.01]	0.000
Hit popularity	0.03	1	988	.870	0.17	0.16	[−1.88, 2.23]	0.000
Jazz (N = 4,300)
Corrected model	152.59	3	4296	<.001
Energy	417.75	1	4296	<.001	−0.16	−20.44	[−0.18, −0.15]	0.089
BPM	6.05	1	4296	.014	−0.01	−2.46	[−0.02, 0.00]	0.001
Hit popularity	0.46	1	4296	.499	1.26	0.68	[−2.39, 4.92]	0.000
Latin (N = 633)
Corrected model	113.11	3	629	<.001
Energy	337.22	1	629	<.001	−0.11	−18.36	[−0.12, −0.10]	0.349
BPM	2.27	1	629	.132	−0.01	−1.51	[−0.02, 0.00]	0.004
Hit popularity	0.90	1	629	.342	0.87	0.95	[−0.93, 2.68]	0.001
Pop (N = 58,250)
Corrected model	8648.64	3	58246	<.001
Energy	25121.46	1	58246	<.001	−0.10	−158.50	[−0.10, −0.09]	0.301
BPM	188.08	1	58246	<.001	−0.01	−13.71	[−0.01, −0.01]	0.003
Hit popularity	0.44	1	58246	.507	0.03	0.66	[−0.07, 0.13]	0.000
Rap or hip hop (N = 8,296)
Corrected model	210.19	3	8292	<.001
Energy	616.20	1	8292	<.001	−0.03	−24.82	[−0.03, −0.02]	0.069
BPM	7.52	1	8292	.006	0.00	−2.74	[−0.01, 0.00]	0.001
Hit popularity	4.72	1	8292	.030	−0.16	−2.17	[−0.31, −0.02]	0.001
Reggae or ska (N = 215)
Corrected model	18.47	3	211	<.001
Energy	42.83	1	211	<.001	−0.11	−6.54	[−0.14, −0.08]	0.169
BPM	6.08	1	211	.014	−0.03	−2.47	[−0.05, −0.01]	0.028
Hit popularity	1.91	1	211	.169	−13.55	−1.38	[−32.89, 5.78]	0.009
Rock (N = 44,307)
Corrected model	2833.91	3	44303	<.001
Energy	8013.60	1	44303	<.001	−0.04	−89.52	[−0.04, −0.04]	0.153
BPM	184.57	1	44303	<.001	−0.01	−13.59	[−0.01, −0.01]	0.004
Hit popularity	37.69	1	44303	<.001	0.57	6.14	[0.39, 0.75]	0.001
Soul or R&B (N = 869)
Corrected model	53.82	3	865	<.001
Energy	160.16	1	865	<.001	−0.09	−12.66	[−0.11, −0.08]	0.156
BPM	0.31	1	865	.577	0.00	0.56	[−0.01, 0.01]	0.000
Hit popularity	0.16	1	865	.691	−0.75	−0.40	[−4.45, 2.95]	0.000
Soundtracks (N = 406)
Corrected model	10.96	3	402	<.001
Energy	27.93	1	402	<.001	−0.09	−5.29	[−0.13, −0.06]	0.065
BPM	7.28	1	402	.007	−0.01	−2.70	[−0.02, 0.00]	0.018
Hit popularity	0.63	1	402	.429	−3.80	−0.79	[−13.24, 5.63]	0.002
World (N = 542)
Corrected model	66.30	3	538	<.001
Energy	197.04	1	538	<.001	−0.10	−14.04	[−0.11, −0.08]	0.268
BPM	2.51	1	538	.113	−0.01	−1.59	[−0.02, 0.00]	0.005
Hit popularity	0.03	1	538	.867	−0.50	−0.17	[−6.37, 5.36]	0.000

DF = degrees of freedom; CI = confidence interval; BPM = beats per minute.

Table 2a.

Means, Standard Errors, 95% Confidence Intervals, and Deviation Contrasts for the GLMM Analysis Concerning Genre Predicting Mood 1.

Genre label	M	SE	95% CI	Deviation contrast: Genre compared with the mean
Genre label	M	SE	95% CI	t	p	95% CI	η²
Alternative or indie	4.56	0.18	[4.21, 4.90]	−18.00	<.001	[−3.54, −2.84]	0.002
Children’s	5.85	0.57	[4.73, 6.98]	−3.48	.001	[−2.96, −0.83]	0.000
Christian or gospel	7.37	0.33	[6.71, 8.02]	−1.19	.236	[−1.01, 0.25]	0.000
Classical or opera	31.62	0.07	[31.48, 31.76]	252.92	<.001	[23.69, 24.06]	0.256
Country	4.60	0.10	[4.40, 4.79]	−27.77	<.001	[−3.37, −2.93]	0.004
Electronica or dance	3.33	0.04	[3.25, 3.41]	−58.63	<.001	[−4.56, −4.27]	0.018
Folk	5.20	0.16	[4.89, 5.51]	−15.69	<.001	[−2.86, −2.23]	0.001
Jazz	9.79	0.08	[9.64, 9.94]	21.11	<.001	[1.86, 2.24]	0.002
Latin	2.78	0.20	[2.39, 3.17]	−25.21	<.001	[−5.35, −4.58]	0.003
New age	25.35	0.66	[24.06, 26.64]	28.30	<.001	[16.39, 18.83]	0.004
Pop	3.56	0.02	[3.52, 3.60]	−61.17	<.001	[−4.32, −4.05]	0.020
Rap or hip hop	2.71	0.06	[2.60, 2.82]	−60.43	<.001	[−5.20, −4.87]	0.019
Reggae or ska	1.98	0.34	[1.31, 2.64]	−17.73	<.001	[−6.41, −5.13]	0.002
Rock	2.16	0.02	[2.11, 2.20]	−80.65	<.001	[−5.73, −5.45]	0.034
Soul or R&B	1.42	0.17	[1.09, 1.75]	−36.85	<.001	[−6.66, −5.99]	0.007
Soundtracks	12.89	0.25	[12.41, 13.37]	21.35	<.001	[4.67, 5.62]	0.002
World	6.51	0.21	[6.09, 6.92]	−5.87	<.001	[−1.65, −0.83]	0.000

Note. SE = standard error; CI = confidence interval.

F(16, 143336) = 1617.47, p < .001, η_p² = .153, overall mean = 7.75.

Table 2b.

Means, Standard Errors, 95% Confidence Intervals, and Deviation Contrasts for the GLMM Analysis Concerning Genre Predicting Mood 2.

Genre label	M	SE	95% CI	Deviation contrast: Genre compared with the mean
Genre label	M	SE	95% CI	t	p	95% CI	η²
Alternative or indie	8.21	0.24	[7.75, 8.68]	−22.05	<.001	[−5.75, −4.81]	0.003
Children’s	18.56	0.78	[17.04, 20.08]	6.90	<.001	[3.63, 6.51]	0.000
Christian or gospel	12.72	0.45	[11.83, 13.60]	−1.79	.073	[−1.63, 0.07]	0.000
Classical or opera	6.89	0.10	[6.70, 7.08]	−51.74	<.001	[−6.85, −6.35]	0.014
Country	16.75	0.13	[16.49, 17.01]	21.24	<.001	[2.96, 3.56]	0.002
Electronica or dance	7.67	0.05	[7.56, 7.77]	−57.27	<.001	[−6.03, −5.63]	0.017
Folk	16.94	0.21	[16.53, 17.36]	15.74	<.001	[3.02, 3.88]	0.001
Jazz	13.84	0.10	[13.64, 14.04]	2.64	.008	[0.09, 0.60]	0.000
Latin	19.13	0.27	[18.61, 19.65]	21.18	<.001	[5.12, 6.16]	0.002
New age	7.42	0.89	[5.68, 9.17]	−7.22	<.001	[−7.72, −4.42]	0.000
Pop	15.15	0.03	[15.09, 15.20]	17.90	<.001	[1.48, 1.84]	0.002
Rap or hip hop	11.85	0.07	[11.70, 11.99]	−14.62	<.001	[−1.87, −1.43]	0.001
Reggae or ska	23.01	0.46	[22.12, 23.91]	21.65	<.001	[8.66, 10.38]	0.003
Rock	9.70	0.03	[9.63, 9.76]	−40.53	<.001	[−3.98, −3.61]	0.009
Soul or R&B	17.85	0.23	[17.40, 18.30]	18.80	<.001	[3.90, 4.81]	0.002
Soundtracks	7.27	0.33	[6.62, 7.93]	−19.10	<.001	[−6.86, −5.58]	0.002
World	16.42	0.29	[15.86, 16.99]	10.27	<.001	[2.37, 3.49]	0.001

Note. SE = standard error; CI = confidence interval.

F(16, 143335) = 2014.14, p < .001, η_p² = .184, overall mean = 13.49.

Table 2c.

Means, Standard Errors, 95% Confidence Intervals, and Deviation Contrasts for the GLMM Analysis Concerning Genre Predicting Mood 3.

Genre label	M	SE	95% CI	Deviation contrast: Genre compared with the mean
Genre label	M	SE	95% CI	t	p	95% CI	η²
Alternative or indie	25.68	0.40	[24.90, 26.47]	37.69	<.001	[14.51, 16.10]	0.008
Children’s	3.89	1.32	[1.31, 6.47]	−5.20	<.001	[−8.92, −4.04]	0.000
Christian or gospel	8.12	0.77	[6.62, 9.62]	−3.08	.002	[−3.70, −0.82]	0.000
Classical or opera	6.92	0.17	[6.60, 7.24]	−15.97	<.001	[−3.88, −3.03]	0.001
Country	9.25	0.23	[8.80, 9.69]	−4.34	<.001	[−1.64, −0.62]	0.000
Electronica or dance	6.66	0.09	[6.49, 6.84]	−21.52	<.001	[−4.05,−3.38]	0.002
Folk	9.65	0.36	[8.94, 10.36]	−1.96	.050	[−1.46, 0.00]	0.000
Jazz	4.14	0.17	[3.80, 4.48]	−28.06	<.001	[−6.67, −5.80]	0.004
Latin	8.97	0.45	[8.08, 9.86]	−3.11	.002	[−2.29, −0.52]	0.000
New age	4.21	1.51	[1.25, 7.17]	−4.32	<.001	[−8.96,−3.37]	0.000
Pop	14.57	0.05	[14.48, 14.63]	26.73	<.001	[3.89, 4.50]	0.004
Rap or hip hop	5.21	0.13	[4.97, 5.46]	−27.03	<.001	[−5.54, −4.79]	0.004
Reggae or ska	2.66	0.78	[1.14, 4.19]	−10.35	<.001	[−9.18, −6.25]	0.001
Rock	33.74	0.05	[33.63, 33.85]	147.04	<.001	[23.05, 23.68]	0.104
Soul or R&B	11.86	0.39	[11.11, 12.62]	3.79	<.001	[0.72, 2.26]	0.000
Soundtracks	12.16	0.57	[11.05, 13.27]	3.23	.001	[0.70, 2.87]	0.000
World	8.69	0.49	[7.73, 9.65]	−3.49	<.001	[−2.64, −0.74]	0.000

Note. SE = standard error; CI = confidence interval.

F(16, 143335) = 8190.39, p < .001, η_p² = .478, overall mean = 10.38.

Table 2d.

Means, Standard Errors, 95% Confidence Intervals, and Deviation Contrasts for the GLMM Analysis Concerning Genre Predicting Mood 4.

Genre label	M	SE	95% CI	Deviation contrast: Genre compared with the mean
Genre label	M	SE	95% CI	t	p	95% CI	η²
Alternative or indie	8.48	0.22	[8.04, 8.91]	−22.32	<.001	[−5.48, −4.60]	0.003
Children’s	24.40	0.73	[22.97, 25.83]	15.72	<.001	[9.53, 12.24]	0.001
Christian or gospel	15.39	0.43	[14.55, 16.22]	4.59	<.001	[1.07, 2.67]	0.000
Classical or opera	12.19	0.09	[12.01, 12.37]	−11.02	<.001	[−1.56, −1.09]	0.001
Country	13.50	0.13	[13.25, 13.75]	−0.11	.916	[−0.30, 0.27]	0.000
Electronica or dance	10.27	0.05	[10.17, 10.37]	−33.81	<.001	[−3.43, −3.06]	0.006
Folk	13.79	0.20	[13.39, 14.18]	1.32	.187	[−0.13, 0.68]	0.000
Jazz	23.33	0.10	[23.14, 23.52]	79.51	<.001	[9.58, 10.06]	0.033
Latin	13.36	0.25	[12.87, 13.85]	−0.62	.535	[−0.65, 0.34]	0.000
New age	12.00	0.84	[10.36, 13.65]	−1.91	.056	[−3.07, 0.04]	0.000
Pop	10.75	0.03	[10.69, 10.80]	−31.74	<.001	[−2.94, −2.60]	0.005
Rap or hip hop	12.61	0.07	[12.48, 12.75]	−8.50	<.001	[−1.11, −0.69]	0.000
Reggae or ska	14.95	0.43	[14.10, 15.80]	3.46	.001	[0.62, 2.25]	0.000
Rock	6.24	0.03	[6.19, 6.30]	−82.30	<.001	[−7.44, −7.10]	0.035
Soul or R&B	14.38	0.22	[13.96, 14.81]	3.98	<.001	[0.44, 1.30]	0.000
Soundtracks	7.14	0.32	[6.52, 7.76]	−20.76	<.001	[−6.98, −5.77]	0.002
World	16.97	0.27	[16.44, 17.51]	12.85	<.001	[2.93, 3.99]	0.001

Note. SE = standard error; CI = confidence interval.

F(16, 143335) = 2536.27, p < .001, η_p² = .221, overall mean = 13.51.

Table 2e.

Means, Standard Errors, 95% Confidence Intervals, and Deviation Contrasts for the GLMM Analysis Concerning Genre Predicting Mood 5.

Genre label	M	SE	95% CI	Deviation contrast: Genre compared with the mean
Genre label	M	SE	95% CI	t	p	95% CI	η²
Alternative or indie	13.23	0.33	[12.59, 13.87]	−5.49	<.001	[−2.46, −1.17]	0.000
Children’s	15.00	1.07	[12.90, 17.10]	−0.04	.967	[−2.03, 1.95]	0.000
Christian or gospel	11.73	0.62	[10.51, 12.95]	−5.54	<.001	[−4.48, −2.14]	0.000
Classical or opera	4.29	0.14	[4.02, 4.55]	−61.08	<.001	[−11.10, −10.41]	0.020
Country	15.48	0.18	[15.12, 15.84]	2.06	.040	[0.02, 0.85]	0.000
Electronica or Dance	15.01	0.07	[14.86, 15.15]	−0.25	.806	[−0.31, 0.24]	0.000
Folk	16.59	0.30	[16.01, 17.17]	5.12	<.001	[0.96, 2.14]	0.000
Jazz	9.93	0.14	[9.66, 10.21]	−28.27	<.001	[−5.46, −4.76]	0.004
Latin	23.50	0.37	[22.78, 24.22]	23.02	<.001	[7.74, 9.18]	0.003
New age	4.91	1.23	[2.50, 7.32]	−8.73	<.001	[−12.40, −7.86]	0.000
Pop	19.98	0.04	[19.90, 20.05]	38.67	<.001	[4.69, 5.19]	0.008
Rap or hip hop	17.25	0.10	[17.06, 17.45]	14.24	<.001	[1.91, 2.52]	0.001
Reggae or ska	20.67	0.63	[19.43, 21.91]	9.27	<.001	[4.43, 6.81]	0.000
Rock	18.23	0.04	[18.15, 18.32]	24.67	<.001	[2.94, 3.44]	0.003
Soul or R&B	25.23	0.32	[24.61, 25.84]	31.83	<.001	[9.56, 10.81]	0.005
Soundtracks	8.80	0.46	[7.90, 9.71]	−13.88	<.001	[−7.12, −5.36]	0.001
World	15.89	0.40	[15.11, 16.67]	2.15	.032	[0.07, 1.62]	0.000

Note. SE = standard error; CI = confidence interval.

F(16, 143335) = 1234.87, p < .001, η_p² = .121, overall mean = 15.04.

Table 2f.

Means, Standard Errors, 95% Confidence Intervals, and Deviation Contrasts for the GLMM Analysis Concerning Genre Predicting Mood 6.

Genre label	M	SE	95% CI	Deviation contrast: Genre compared with the mean
Genre label	M	SE	95% CI	t	p	95% CI	η²
Alternative or indie	9.25	0.25	[8.75, 9.74]	−7.92	<.001	[−2.50, −1.51]	0.000
Children’s	14.60	0.82	[12.99, 16.21]	4.31	<.001	[1.82, 4.87]	0.000
Christian or gospel	15.77	0.48	[14.83, 16.71]	9.86	<.001	[3.62, 5.42]	0.001
Classical or opera	9.56	0.10	[9.35, 9.76]	−12.56	<.001	[−1.96, −1.43]	0.001
Country	19.38	0.14	[19.10, 19.66]	50.09	<.001	[7.81, 8.45]	0.013
Electronica or dance	2.79	0.06	[2.68, 2.90]	−78.60	<.001	[−8.68, −8.26]	0.032
Folk	16.22	0.23	[15.78, 16.66]	21.39	<.001	[4.51, 5.42]	0.002
Jazz	18.44	0.11	[18.28, 18.65]	51.84	<.001	[6.92, 7.46]	0.014
Latin	10.12	0.28	[9.56, 10.67]	−4.03	<.001	[−1.69, −0.58]	0.000
New age	15.77	0.94	[13.93, 17.62]	5.08	<.001	[2.77, 6.26]	0.000
Pop	9.78	0.03	[9.73, 9.84]	−15.01	<.001	[−1.66, −1.28]	0.001
Rap or hip hop	3.72	0.08	[3.57, 3.87]	−63.19	<.001	[−7.77, −7.30]	0.021
Reggae or ska	10.93	0.49	[9.98, 11.88]	−0.69	0.488	[−1.24, 0.59]	0.000
Rock	5.59	0.03	[5.52, 5.65]	−57.14	<.001	[−5.86, −5.47]	0.017
Soul or R&B	7.82	0.24	[7.34, 8.29]	−14.00	<.001	[−3.92, −2.96]	0.001
Soundtracks	10.19	0.35	[9.49, 10.88]	−3.10	0.002	[−1.74, −0.39]	0.000
World	11.39	0.31	[10.79, 11.99]	0.46	0.647	[−0.45, 0.73]	0.000

Note. SE = standard error; CI = confidence interval.

F(16, 143335) = 2394.97, p < .001, η_p² = .211, overall mean = 11.25.

Discussion

Energy, BPM, Hit Popularity, and Mood (Hypothesis 1)

Hypothesis 1 addressed the arousal dimension of the circumplex. Table 1(a) to (f) shows the relationship between each of energy, BPM, and hit popularity for each of the six moods in the case of both the overall dataset and for each genre in turn. Across the dataset as a whole, energy was related negatively to Moods 1 (clean, simple, and relaxing), 4 (mystery, luxury, and comfort), and 6 (calm, peace, and tranquility) and positively to Moods 3 (passion, romance, and power) and 5 (energetic, bold, and outgoing). With very few exceptions, the same direction of (significant) findings was also identified for each of these moods in the case of each of the genres considered. On the whole, therefore, the results concerning energy appear consistent with the circumplex model. Findings concerning energy and Mood 2 (happy, hopeful, and ambition) were, however, more mixed: Although the relationship was negative in the overall dataset, results concerning several of the individual genres indicated a positive relationship. One possible explanation of this is that Mano (1991) and Russell and Mehrabian (1977) have shown that the adjectives associated with Mood 2 sit around the midway point of the activity dimension of the circumplex (although whether they are more prone to this issue than are the other moods investigated here is debatable).

As expected, the corresponding results concerning BPM yielded much weaker effect sizes, although many of the individual tests were nonetheless significant at the restricted alpha level, which is itself pleasing given that BPM is only one factor that contributes to the overall arousal of a piece. Across the dataset as a whole, BPM was related positively to Mood 3 (passion, romance, and power), and negatively to Moods 4 (mystery, luxury, and comfort) and 6 (calm, peace, and tranquility), all of which is consistent with the circumplex model. Given the small effect sizes in the overall dataset, it is unsurprising, therefore, that only some of the individual genres yielded associations between BPM and the six mood scores, although again those that were significant were usually in the direction predicted by the circumplex model (although again subject to low effect sizes). There were negative relationships between Mood 1 (clean, simple, and relaxing) and BPM for jazz and pop but also a positive relationship for electronica or dance. There were positive relationships between Mood 2 (happy, hopeful, and ambition) and BPM for country, jazz, and pop but also a negative relationship for electronica or dance and rap or hip hop. There were positive relationships between Mood 3 (passion, romance, and power) and BPM for alternative or indie, country, jazz, pop, and rock. There were negative relationships between Mood 4 (mystery, luxury, and comfort) and BPM for alternative or indie, country, electronica or dance, pop, rap or hip hop, and rock. There were positive relationships between Mood 5 (energetic, bold, and outgoing) and BPM for jazz and pop but also a negative relationship for electronica or dance. There were negative relationships between Mood 6 (calm, peace, and tranquility) and BPM for alternative or indie, electronica or dance, pop, and rock. In general, the results support Hypothesis 1.

Mood and Commercial Success (Hypothesis 2)

Hypothesis 2 addressed the pleasantness dimension of the circumplex. As anticipated, although there were several significant relationships between hit popularity and the six moods, Table 1(a) to (f) indicates that the nature of these were not consistent with findings concerning the pleasantness dimension of the circumplex, and so do not support Hypothesis 2. We were less confident that the results would satisfy this second hypothesis, however. Recent findings have described the importance of distinguishing the emotions evoked by music from the affective valence of these emotions, such that, for instance, one might regard a piece of music as distressing but enjoy that music as a direct consequence of this sadness. In a direct test of this, Schubert (2013) asked participants to select music that they loved and music that they hated, with analyses showing that many participants selected as liked music that which evoked negative emotions such as sadness and grief (and note that Hanich, Wagner, Shah, Jacobsen, & Menninghaus, 2014, make similar arguments in the light of data concerning participants’ responses to sad films): Schubert argued that, in instances such as these, the emotion valence is of course negative, but crucially that the affective response is separate and positively valenced. Within this framework, a piece of music regarded as exciting would likely have both a positive emotional valence and a positive affective valence; a piece regarded as boring would likely have both a negative emotional valence and a negative affective valence; but a piece that is enjoyed because it evokes sadness and grief, or any other emotion typically located in the lower half of the pleasantness dimension, would have a negative emotional valence but nonetheless also have a positive affective valence.

Similar fundamental arguments are made by Sachs, Damasio, and Habibi’s (2015) review of the persistent popularity of sad music, which argues that this is pleasurable because it serves a quasi-homeostatic function. They describe the results of several psychological and neuroimaging studies indicating that sad music evokes pleasure if it is nonthreatening, aesthetically pleasing, and has positive psychological effects (e.g., evocation of empathy, nostalgia, or other specific and desired moods). Of course, this mechanism is not mutually exclusive of Schubert’s, such that the latter describes arguably the same phenomena in psychological and conceptual terms, whereas Sachs et al.’s account has a clearer physiological emphasis.

Whichever of these explanations is more accurate, both have the same implication that appears consistent with the present findings. When the circumplex relates pleasantness to the more specific emotional connotations of that music, the approach arguably underspecifies both concepts: Specifically, it conflates the emotional and affective valence of a person’s response to the music, such that the latter might rely upon an idiosyncratic, cognitive component that is subject to wide-ranging individual differences. The same argument applies also to the use of sales data in the present research as a proxy for the pleasantness dimension. All these arguments notwithstanding, however, even if one questions the validity of the pleasantness dimension of the circumplex (or of sales data as a proxy for the pleasantness dimension) as a true measure of the valence of a particular affective response, this aspect of the present dataset also allows us to address a different question of considerable practical relevance, namely the potential correlation between music sales and the expression of certain emotions: Across all music of any commercial relevance in the United Kingdom, the research can determine which musical emotions are most popular.

In the light of this argument, there are three interpretations of the results concerning Hypothesis 2. The first is that the measure is a valid representation of the pleasantness dimension of the circumplex and that the latter is not related to emotion as predicted. The second is that the moods employed in the research (which were, in effect, determined by the music industry) do not represent a full range of states along the continuum of the valence dimension of the circumplex. The third is that hit popularity is not an adequate representation of the pleasantness dimension of the circumplex. Of these explanations we favor the latter two, and particularly the third, for reasons set out immediately above. As such, it may well be crass to argue that the current measure of hit popularity truly captures the pleasantness dimension of the circumplex or the emotional and affective valence of responses to the music: Neither, of course, do the present results provide strong support for the pleasantness dimension of the circumplex model.

Nonetheless, the relationships that do exist between hit popularity and mood do provide a fascinating insight into the emotional connotations of pieces that enjoy greater commercial success. Although the effect sizes were very small, the overall dataset shows significant, positive relationships between hit popularity and each of Moods 1 (clean, simple, and relaxing), 4 (mystery, luxury, and comfort), and 6 (calm, peace, and tranquility); but negative relationships between hit popularity and each of Moods 2 (happy, hopeful, and ambition), 3 (passion, romance, and power), and 5 (energetic, bold, and outgoing), such that the former moods are associated with greater commercial success and the latter moods are associated with lower commercial success. Of all these findings, it is particularly interesting that Mood 2 (happy, hopeful, and ambition) was associated negatively with commercial success, despite the caricature that sales charts and commercial radio airplay are dominated by emotionally upbeat music; and that Mood 4 (mystery, luxury, and comfort) demonstrated the strongest positive association with commercial success, and Mood 5 (energetic, bold, and outgoing) demonstrated the strongest negative association with commercial success.

However, these patterns in the overall dataset mask several variations between genres, such that commercial success in one genre appears to require evocation of different moods compared with other genres: More explicitly, the emotion-based criteria of commercial success vary between genres. Mood 1 (clean, simple, and relaxing) was associated positively with commercial success in the cases of classical music, electronica or dance, pop, rock, and soul or R&B. Mood 2 (happy, hopeful, and ambition) was associated negatively with commercial success in the case of classical music, electronica or dance, pop, and rock. Mood 3 (passion, romance, and power) was associated positively with commercial success in the case of electronica or dance and was associated negatively with commercial success in the case of rock. Mood 4 (mystery, luxury, and comfort) was associated positively with commercial success in the case of pop and rock and negatively with commercial success in the case of alternative or indie and classical music. Mood 5 (energetic, bold, and outgoing) was associated negatively with commercial success in the case of country, pop, rock, and soul or R&B. Mood 6 (calm, peace, and tranquility) was associated positively with commercial success in the case of rock and negatively with commercial success in the case of classical music.

Genre and Mood

This in turn leads to the subsidiary issue investigated on an exploratory basis by the present research, namely differences between genres in mood. Table 2(a) to (f) indicates a very large number of differences between genres in the moods they connote. For the sake of space, we hesitate to enter into a detailed description of the moods evoked by each genre and where each significant difference lies. However, for the sake of illustration, consider the findings concerning the alternative or indie genre as shown in Table 2(a) to (c). The mean percentage score was 4.56 for Mood 1 (clean, simple, and relaxing), 8.21 for Mood 2 (happy, hopeful, and ambition), and 25.68 for Mood 3 (passion, romance, and power), such that alternative or indie music is not very reflective of Mood 1 or 2, and much more likely to convey Mood 3 (passion, romance, and power) than it is to convey the other moods. In short, different genres are associated with different moods to differing extents, and this has clear implications for those wishing to use music genre as a means of influencing mood either in personal, everyday music usage, given recent research showing the importance of perceived control over the music (Krause et al., 2014); therapeutic settings in which music has health-related effects that are contingent upon reliable induction of mood (Standley, 1995); or in commercial contexts, such as the use of music in advertising or in-store to influence consumers’ moods and in turn various aspects of their purchasing behaviors (North & Hargreaves, 2008). The present findings might also provide useful guidance for future work in public health and criminology that has identified elevated mental health problems and juvenile offending among those who listen to certain musical styles, particularly rock and rap: It is noteworthy in this context that Table 2(a) to (f) shows that rap or hip hop and rock scored lowest of the musical styles on Moods 1 (clean, simple, and relaxing) and 6 (calm, peace, and tranquility). Also interesting in this context, however, is that classical music scored much lower than the other genres on Mood 2 (happy, hopeful, and ambition), which may illustrate why the public health research shows associations between musical taste and mental health that are not exclusive to rap and rock music (see e.g., Stack’s, 2002, evidence concerning suicide acceptance in opera audiences).

Limitations

One of the clear advantages of the archival approach adopted here is the potential to test theory using a very large sample of music and sales information from entire populations. However, inherent to the approach are a number of limitations which deserve attention. First, we have briefly mentioned already the difficulty of testing the pleasantness dimension of the circumplex via archival data. Specifically, while sales charts and radio airplay can provide a population-wide measure of the overall popularity of a given piece, there is an issue with the failure of this measure to distinguish between emotional and affective valence. More fine-grained measures of these two variables, which include reactions to music at the negative end of the pleasantness dimension, will need to be developed before this aspect of the circumplex model can be tested meaningfully through means such as those employed here. In terms of their ability to speak to the circumplex model, we have much more confidence in conclusions drawn from the present data concerning energy than we do in those concerning pleasantness or chart performance.

Second, as with much of the research on music and emotion, the present methodology is unable to account for any individual differences in emotional reactions to music, and in particular those arising from extrinsic associations that a given piece has for a given listener (or for entire populations through the use of the music in question in, for instance, advertising campaigns). In a similar vein, the current approach to data collection cannot account for the impact of the location of listening on emotional response, despite numerous recent studies associating the two (e.g., Krause, North, & Hewitt, 2016).

Finally, the database of music analyzed was limited to that which had enjoyed popularity in the United Kingdom, such that the present findings cannot speak to music and emotion in other cultures. However, although the findings concerning genre and mood would likely differ cross-culturally, we are optimistic that future research concerning energy and mood in even radically different cultures to those investigated here would yield similar findings, given that Russell (1983) found evidence supporting the circumplex among native speakers of Gujurati, Croatian, Japanese, and Chinese; Russell, Lewicka, and Niit (1989) found evidence confirming the circumplex model among Chinese participants; and Furrer, Tjemkes, Aydinlik, and Adolfs (2012) found similar results in Japan.

Conclusion

The present research has found that the mood of a very large sample of music can be predicted by its energy, which is consistent with the circumplex model of affect. Findings concerning BPM and mood were less clear, although the broadly consistent pattern of findings is what might be expected given that the former is clearly just one of several contributors to the overall arousing qualities of music. Findings concerning hit popularity and mood were more equivocal in their support for the circumplex model, although this might be because the measure failed to adequately capture the difference between emotional and affective valence; and the extensive relationships that do exist between hit popularity and mood provide some interesting insights into the preferences of the audiences for differing genres, and how certain genres place more emphasis on certain moods than others. Aside from their theoretical implications for research on the circumplex and aesthetic responses to music, the findings are potentially relevant to music marketing, and perhaps also to a more limited extent to music therapy, marketing, and the public’s everyday music listening habits.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Adrian C. North is a professor of Psychology at Curtin University. His research is on the social and applied psychology of music, with particular interest in aesthetics and the relationship between music listening and subsequent behaviour.

Amanda E. Krause is a research associate in the School of Psychology and Speech Pathology at Curtin University. Her research interests include everyday listening habits, with an emphasis on the relationships between digital music, emerging web technologies, and well-being.

Lorraine P. Sheridan is a forensic psychologist and a senior lecturer in the School of Psychology and Speech Pathology at Curtin University, Western Australia. She has been researching and publishing on stalking since 1997 and provides input on the assessment and management of stalking cases to the police, public figures, and government agencies.

David Ritchie is CEO of Sounds Like Me Limited, a company dedicated to using music-focused AI algorithms concerning personality, music tastes, and listening habits.

References

Alcalde

Ricard

Bonet

Llopis

Marcos

(2008) U.S. Patent No. 20080021851, Washington, DC: U.S. Patent and Trademark Office.

Alcalde

Ricard

Bonet

Llopis

Marcos

(2010) U.S. Patent No. 20100250471, Washington, DC: U.S. Patent and Trademark Office.

Bartlett

D. L.

(1996) Physiological reactions to music and acoustic stimuli. In: Hodges

D. A.

(ed.) Handbook of music psychology, 2nd ed. San Antonio, CA: IMR Press, pp. 343–385.

Berlyne

D. E.

(1971) Aesthetics and psychobiology, New York, NY: Appleton-Century-Crofts.

Bruner

G. C.

(1990) Music, mood, and marketing. Journal of Marketing 54: 94–104.

Carney

D. R.

Colvin

C. R.

(2010) The circumplex structure of affective social behavior. Social Psychological and Personality Science 1(1): 73–80. doi:10.1177/1948550609353135.

Cooke

(1959) The language of music, Oxford, England: Oxford University Press.

Dibben

(2004) The role of peripheral feedback in emotional experience with music. Music Perception 22: 79–115. doi:10.1525/mp.2004.22.1.79.

Eerola, T., Lartillot, O., & Toiviainen, P. (2009, October). Prediction of multidimensional emotional ratings in music from audio using multivariate regression. Presented at 10th International Society for Music Information Retrieval Conference (ISMIR 2009), Kobe, Japan.

10.

English

Carstensen

L. L.

(2014) Emotional experience in the mornings and the evenings: Consideration of age differences in specific emotions by time of day. Frontiers in Psychology 5: 185, doi:10.3389/fpsyg.2014.00185.

11.

Furrer

Tjemkes

B. V.

Aydinlik

A. U.

Adolfs

(2012) Responding to adverse situations within exchange relationships: The cross-cultural validity of a circumplex model. Journal of Cross-Cultural Psychology 43: 943–966. doi:10.1177/0022022111415671.

12.

Gabrielsson

Juslin

P. N.

(1996) Emotional expression in music performance: Between the performer’s intention and the listener’s experience. Psychology of Music 24: 68–91.

13.

Gabrielsson

Lindström

(2001) The influence of musical structure on emotional expression. In: Juslin

P. N.

Sloboda

J. A.

(eds) Music and emotion: Theory and research, Oxford, England: Oxford University Press, pp. 223–248.

14.

Hanich

Wagner

Shah

Jacobsen

Menninghaus

(2014) Why we like to watch sad films: The pleasure of being moved in aesthetic experiences. Psychology of Aesthetics, Creativity, and the Arts 8: 130–143.

15.

Hevner

(1936) Experimental studies of the elements of expression in music. American Journal of Psychology 48: 246–268.

16.

Juslin

P. N.

(2000) Cue-utilisation in communication of emotion in music performance: Relation performance to perception. Journal of Experimental Psychology 26: 1797–1813.

17.

Juslin

P. N.

(2005) From mimesis to catharsis: Expression, perception, and induction of emotion in music. In: Miell

MacDonald

R. A. R.

Hargreaves

D. J.

(eds) Musical communication, Oxford, England: Oxford University Press, pp. 85–115.

18.

Juslin

P. N.

Laukka

(2000) Improving emotional communication in music performance through cognitive feedback. Musicae Scientiae 4: 151–183.

19.

Juslin

P. N.

Laukka

(2003) Communication of emotion in vocal expression and music performance: Different channels, same code? Psychological Bulletin 129(5): 770–814. doi:10.1037/0033-2909.129.5.770.

20.

Kaminska

Woolf

(2000) Melodic line and emotion: Cooke’s theory revisited. Psychology of Music 28(2): 133–153. doi:10.1177/0305735600282003.

21.

Khalfa

Peretz

Blondin

J.-P.

Manon

(2002) Event-related skin conductance responses to musical emotions in humans. Neuroscience Letters 328(2): 145–149. doi:10.1016/S0304-3940(02)00462-7.

22.

Krause

A. E.

North

A. C.

Hewitt

L. Y.

(2014) Music selection behaviors in everyday listening. Journal of Broadcasting and Electronic Media 58(2): 306–323. doi:10.1080/08838151.2014.906437.

23.

Krause

A. E.

North

A. C.

Hewitt

L. Y.

(2016) The role of location in everyday experiences of music. Psychology of Popular Media Culture 5(3): 232–257.

24.

Kreutz

Ott

Teichmann

Osawa

Vaitl

(2008) Using music to induce emotions: Influences of musical preference and absorption. Psychology of Music 36(1): 101–126. doi:10.1177/0305735607082623.

25.

Laurier

Sordo

Serra

Herrera

(2009) Music mood representations from social tags. In: Hirata

Tzanetakis

Yoshii

(eds) Proceedings of the 10th International Society for Music Information Retrieval Conference, Kobe, Japan: ISMIR, pp. 381–386.

26.

Leviston

Price

Bishop

(2014) Imagining climate change: The role of implicit associations and affective psychological distancing in climate change responses. European Journal of Social Psychology 44(5): 441–454. doi:10.1002/ejsp.2050.

27.

Loizou

Karageorghis

C. I.

Bishop

D. T.

(2014) Interactive effects of video, priming, and music on emotions and the needs underlying intrinsic motivation. Psychology of Sport and Exercise 15(6): 611–619. doi:10.1016/j.psychsport.2014.06.009.

28.

Madsen

C. K.

(1998) Emotion versus tension in Haydn’s Symphony no. 104 as measured by the two-dimensional continuous response digital interface. Journal of Research in Music Education 46: 546–554.

29.

Mano

(1991) The structure and intensity of emotional experiences: Method and context convergence. Multivariate Behavioral Research 26: 389–411.

30.

McFarland

R. A.

(1985) Relationship of skin temperature changes to the emotions accompanying music. Biofeedback and Self-Regulation 10: 255–267.

31.

North

A. C.

Hargraves

D. J.

(1997) Liking, arousal potential, and the emotions expressed by music. Scandinavian Journal of Psychology 38: 45–53. doi:10.1111/1467–9450.00008.

32.

North

A. C.

Hargreaves

D. J.

(2008) The social and applied psychology of music, Oxford, England: Oxford University Press.

33.

North

A. C.

Shilcock

Hargreaves

D. J.

(2003) The effect of musical style on restaurant customers’ spending. Environment and Behavior 35(5): 712–718. doi:10.1177/0013916503254749.

34.

Nyklicek

Thayer

J. F.

van Doornen

L. J. P.

(1997) Cardiorespiratory differentiation of musically-induced emotions. Journal of Psychophysiology 11: 304–321.

35.

Panksepp

Bekkedal

M. Y. V.

(1997) The affective cerebral consequence of music: Happy vs. sad effects on the EEG and clinical implications. International Journal of Arts Medicine 5: 18–27.

36.

Posner

Russell

J. A.

Gerber

Gorman

Colibazzi

Peterson

B. S.

(2009) The neurophysiological bases of emotion: An fMRI study of the affective circumplex using emotion-denoting words. Human Brain Mapping 30(3): 883–895. doi:10.1002/hbm.20553.

37.

Qin

Zheng

Tian

Zheng

(2014) An emotion-oriented music recommendation algorithm fusing rating and trust. International Journal of Computational Intelligence Systems 7: 371–381.

38.

Rickard

N. S.

(2004) Intense emotional responses to music: A test of the physiological arousal hypothesis. Psychology of Music 32(4): 371–388. doi:10.1177/0305735604046096.

39.

Ritossa

D. A.

Rickard

N. S.

(2004) The relative utility of ‘pleasantness and liking’ dimensions in predicting the emotions expressed by music. Psychology of Music 32(1): 5–22. doi:10.117/0305735604039281.

40.

Russell

J. A.

(1978) Evidence of convergent validity on the dimensions of affect. Journal of Personality and Social Psychology 36: 1152–1168.

41.

Russell

J. A.

(1983) Pancultural aspects of the human conceptual organization of emotions. Journal of Personality and Social Psychology 45(6): 1281–1288.

42.

Russell

J. A.

Lewicka

Niit

(1989) A cross-cultural study of a circumplex model of affect. Journal of Personality and Social Psychology 57: 848–856.

43.

Russell

J. A.

Mehrabian

(1977) Evidence for a three-factor theory of emotions. Journal of Research in Personality 11: 273–294.

44.

Saari

Eerola

(2013) Semantic computing of moods based on tags in social media of music. IEEE Transactions on Knowledge and Data Engineering 26: 2548–2560. doi:10.1109/TKDE.2013.128.

45.

Saari

Eerola

Fazekas

Barthet

Lartillot

Sandler

(2013) The role of audio and tags in music mood prediction: A study using semantic layer projection. International Society for Music Information Retrieval Conference 2013: 201–206.

46.

Sachs

M. E.

Damasio

Habibi

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

47.

Scherer

K. R.

Zentner

M. R.

(2001) Emotional effects of music: Production rules. In: Juslin

P. N.

Sloboda

J. A.

(eds) Music and emotion: Theory and research, Oxford, England: Oxford University Press, pp. 361–392.

48.

Schubert

(2004) Modeling perceived emotion with continuous musical features. Music Perception 21(4): 561–585. doi:10.1525/mp.2004.21.4.561.

49.

Schubert

(2013) Loved music can make a listener feel negative emotions. Musicae Scientiae 17: 11–26. doi: 10.1177/1029864912461321.

50.

Scirea, M., Nelson, M. J., & Togelius, J. (2015). Moody music generator: Characterising control parameters using crowdsourcing. Paper presented at the 4th Conference on Evolutionary and Biologically Inspired Music, Sound, Art and Design, Copenhagen, Denmark.

51.

Sloboda

J. A.

Juslin

P. N.

(2001) Psychological perspectives on music and emotion. In: Juslin

P. N.

Sloboda

J. A.

(eds) Music and emotion, Oxford, England: Oxford University Press, pp. 71–104.

52.

Stack

(2000) Blues fans and suicide acceptability. Death Studies 24(3): 223–231. doi:10.1080/074811800200559.

53.

Stack

(2002) Opera subculture and suicide for honor. Death Studies 26(5): 431–437. doi:10.1080/07481180290086763.

54.

Stack

Gundlach

J. H.

(1992) The effect of country music on suicide. Social Forces 71(1): 211–218.

55.

Standley

(1995) Music as a therapeutic intervention in medical and dental treatment: Research and clinical applications. In: Wigram

Saperstone

West

(eds) The art and science of music therapy: A handbook, New York, NY: Routledge, pp. 3–22.

56.

Straehley

I. C.

Loebach

J. L.

(2014) The influence of mode and musical experience on the attribution of emotions to melodic sequences. Psychomusicology 24(1): 21–34. doi:10.1037/pmu0000032.

57.

Tseng

Bansal

Liu

Gerber

A. J.

Goh

Posner

Peterson

B. S.

(2014) Using the circumplex model of affect to study valence and arousal ratings of emotional faces by children and adults with autism spectrum disorders. Journal of Autism and Developmental Disorders 44(6): 1332–1346. doi:10.1007/s10803-013-1993-6.

58.

Zentner

Eerola

(2010) Self-report measures and models. In: Juslin

P. N.

Sloboda

J. A.

(eds) Handbook of music and emotion: Theory, research, applications, Oxford, UK: Oxford University Press, pp. 187–221.

59.

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.