Can Music Influence Cultural Evaluations? Objective Musical Parameters and Subjective Musical Ratings As Predictors of Cultural Dimensions

Home , Background music, Melodic expectation

Can Music Influence Cultural Evaluations? Objective Musical Parameters and Subjective Musical Ratings as Predictors of Cultural Dimensions

John Melvin G. Treider

Master Thesis in Psychology

Department of Psychology Faculty of Social Sciences

University of Oslo

June 2020

Can Music Influence Cultural Evaluations? Objective Musical Parameters and Subjective Musical Ratings as Predictors of Cultural Dimensions

III

Can Music Influence Cultural Evaluations? Objective Musical Parameters and Subjective Musical Ratings as Predictors of Cultural Dimensions

John Melvin Treider http://www.duo.uio.no/

Acknowledgements

First and foremost, I would like to thank my supervisors, Professor Jonas. R. Kunst and Professor Jonna Vuoskoski, for providing me with indispensable support and feedback whilst writing this thesis. Further, I would like to thank the Neuroband for shared joyful moments and support across peak amplitudes at high frequencies. Finally, I would like to thank RITMO Centre for Interdisciplinary Studies in Rhythm, Time and Motion for funding for this research.

Can Music Influence Cultural Evaluations?

Objective Musical Parameters and Subjective Musical Ratings as Predictors of Cultural Dimensions By John Melvin Treider Supervised by Professor Jonas Kunst and Professor Jonna Vuoskoski Department of Psychology, University of Oslo, Norway

Abstract Resent research suggests that music can convey social intentions through musical parameters (Aucouturier & Canonne, 2017) and modulate implicit cultural attitudes (Vuoskoski et al., 2017). Therefore, musical parameters present in music may partly explain this modulation in attitudes. Nevertheless, it is not yet known whether specific musical parameters are associated with listeners’ cultural evaluations in a consistent manner. Through two studies, this paper attempts to establish a research avenue for investigating the effects of musical parameters on basic dimensions of intergroup evaluation. Study 1 collected survey responses from 52 American participants who were exposed to 30 selected folk-song melodies from different world continents. Linear mixed-effects model analyses tested the influence of objective musical parameters and subjective musical perception on cultural evaluation. Musical parameters and perception consistently predicted cultural evaluations. The most prominent effect of the musical parameters was musical velocity, a measure of pitch onsets. Study 2 attempted to exert more systematic control over musical parameters by altering tempo and dissonance for a subset of the melody excerpts used in Study 1. The study collected responses from 212 American participants who were exposed to 6 folk song melodies in different modulations of tempo and dissonance. Linear mixed- effects model analyses revealed that both dissonance and slow tempo predicted more negative cultural evaluations; most notable was the negative effect of dissonance on cultural warmth. These studies may be used as an outline for future studies aiming to study musical parameters further or investigate the potential of music to bridge different cultures.

Keywords: musical ratings, musical parameters, intergroup evaluations

VII

Table of Contents 1 Introduction……………………………………………………….…………………………1 2 Literature Review……………………………………………………………………………2 2.1 The Contact Hypothesis……………………………….……………………...…....2 2.2 The Indirect Contact Hypothesis………………………….…………….……...….2 2.3 Music as Indirect Intergroup Contact…………………….………………………..3 2.4 Music’s Potential to Influence Cultural Biases………….……………………...…4 2.5 Cultural Evaluation Through Music……………………………………...………..5 2.5.1 Dimensions Contributing to Cultural Biases ……………………,…...….6 2.5.2 Communicative Aspects of Music……………………………………….6 2.6 Musical Parameters that may Shape Intergroup Biases……...…………………….8 2.6.1 Originality………………………………………………………………..8 2.6.2 Intervals…………………………………………….…………………….9 2.6.3 Modality……………………………………………….…………………9 2.6.4 Tempo…………………….………………….………………………….10 2.6.5 Dissonance……….…………………………………….………………..12 3 The Present Research………………………………………………………….……………17 3.1 Hypotheses……………………………………….……………………………….18 4 Study 1………………………………………………….…………………………………..19 4.1 Methods…………………………….…………………..…………………………19 4.1.1 Participants………………………………………………..…………….20 4.1.2 Song stimuli………………………………………………….…………20 4.1.3 Procedure………………………………………………….……………20 4.1.4 Analyses……………………………………….………………………..21 4.2 Results………………………………………………………….…………………23 4.1.1 Zero-order Correlations………………….……………………………..23 4.1.2 Linear Mixed Models………………….………………..………………24 4.3 Discussion……………………………………….……………………………….28 5 Study 2……………………………………………………………………..……………….32 5.1 Methods…………………………….……………………….……………………32 5.1.1 Participants……………………………………………………………..32 5.1.2 Song Stimuli Selection…………………………………………………32

VIII

5.1.3 Procedure……………………………………………………………….34 5.1.4 Analyses…………………..……………………………………………34 5.2 Results……………………………………………………………………………36 5.2.1 Linear Mixed Models…………………………………………………..36 5.3 Discussion………………………………………………….…………………….40 6 General Discussion………………………………………………………...………………43 6.1 Limitations………………………………………………………...... …………..44 6.2 Future Directions……………………………………...…………...…………….45 7 Conclusion……………………………………………………………………...………….47 8 Funding…………………………………………………………………………...………..47 9 My Role in the Research Project……………………………………….………...………..47 References……………………………………………………………………………....……48 Appendix A…………………………………………………………………………..………63 Appendix B…………………………………………………………………………..………68 Appendix C…………………………………………………………………………..………70

1 Introduction Music is an essential and ever-present part of people’s everyday life (Harwood, 2017). We listen to music both intently and inattentively at concerts, at festivals, at home, and at religious and spiritual assemblies. Moreover, music can subconsciously contribute to creating atmospheres in restaurants, bars, cafes and shopping centers. Music also accompanies scenes in movies, theater plays, television series and commercials, contributing to the induction of a range of different emotions in the listener. In short, music is present everywhere. Music is also an important part of social identity (Lonsdale & North, 2009; Tekman & Hortaçsu, 2002). Almost every country has a national anthem, and many musical styles are related to the unique expression of national pride (Boer et al., 2013; Gilboa & Bodner, 2009). In this way, music is often invoked to bring groups of people together and share expressions of their unique social identity (Bakagiannis & Tarrant, 2006; Lee et al., 2011). Indeed, shared musical preferences have shown to be a great conversation starter, and bonding over musical taste can foster more social attraction between people (Boer et al., 2011; Rentfrow & Gosling, 2006). Furthermore, a much-cited theory in the evolution of music is the social cohesion theory (Roederer, 1984), which proposes that music and dance may have served adaptive functions in an ancestral human environment as they allow for more social bonding and cooperation (Cross & Morley, 2008). There is indeed empirical support for this hypothesis, showing that people who sing together, move together synchronously or tap to the same beat foster more social bonding, cooperation, empathy and liking to each other (Kirschner & Tomasello, 2010; Konvalinka et al., 2010; Kreutz, 2014; Rabinowitch et al., 2013; Tarr et al., 2015). Undoubtedly, music can bring members of a culture closer. Therefore, an interesting research avenue may be to investigate how music operates in promoting bonds between members of different groups and cultures. Perhaps music, as a social, complex artistic expression, may be used to reduce cultural bias and prejudice. However, we know little about the ways in which music may influence the evaluation of cultures. Therefore, through two studies, this thesis will attempt to explore objective and subjective aspects of music that may inform people’s cultural evaluations. Before these studies are presented, the paper will offer a review of relevant previous literature, leading to the rationale of the present research.

2 Literature Review First, this chapter will present literature on increased contact as a proposed method to reduce prejudice and how music may be used as a communicative device to increase contact between cultures. Second, this chapter will examine social and emotional dimensions of intergroup evaluation and how music may inform such dimensions. Finally, the chapter will offer insights on objective musical parameters known to affect musical connotations and preference in music, and how these parameters might influence cultural evaluation.

2.1 The Contact Hypothesis One influential theory of decreasing prejudice between groups is the contact hypothesis, which stems from a suggestion by Allport (1954), who posited that under the appropriate conditions, increasing contact between members of different groups may prove beneficial. Several studies have supported such a method for reducing prejudice (Schmid et al., 2014; Schlueter & Scheepers, 2010; Wagner et al., 2006). Nevertheless, the contact hypothesis has also given rise to the debate about the practical implementation of such extended contact. Even though people are put in situations that may facilitate contact between groups, they may experience more anxiety and feelings of uncertainty and awkwardness in terms of how to behave (Plant & Devine, 2003; Stephan & Stephan, 1985; Stephan et al., 1999). In addition, the relationship between prejudice and contact is bidirectional; prejudice also reduces contact (Binder et al., 2009; Levin et al., 2003). Therefore, even when conditions are economically and structurally optimal for contact to occur, segregation will likely endure (Schuleter et al., 2018). Thus, even though the contact hypothesis has opened up for an interesting avenue of research that is solution-directed, there are still limitations to the theory.

2.2 The Indirect Contact Hypothesis

Several researchers have investigated the role of indirect intergroup contact. This theory posits that communication does not need to derive from direct face-to-face contact between members of different cultures. Rather, contact and communication can occur at many different levels and through various forms of communicative channels. Wright et al. (1997) proposed that simply knowing that an outgroup member is friends with an ingroup member can lead to reduced prejudice, a proposal that has been empirically supported in majority and minority participants from both Europe and America (Zhou et al., 2018). Similarly, observing an interaction between an ingroup and an outgroup member, or simply imagining contact with

an outgroup member may have the same prejudice-reducing effects (Husnu & Crisp, 2010; Mazziotta et al., 2011). Moreover, Schiappa et al. (2005) demonstrated that watching television shows of positive intergroup contact reduced prejudice. Similar findings were found for children’s attitudes towards refugees when reading friendship stories portraying intergroup relations (Cameron et al., 2006). These studies illustrate the immense practical effect of indirect (also referred to as parasocial) contact, especially since arts and mass media have the potential to reach many people simultaneously.

2.3 Music as Indirect Intergroup Contact

The signaling coalition theory (Hagen & Bryant, 2003) employs an evolutionary intergroup perspective to music, positing that music and dance may have been used in the ancestral past as a way of signaling group fitness to other communities, which in turn could enhance the likelihood for survival. Therefore, a group displaying synchronicity and social bonding within the group could indeed be perceived as aggressive defenders or desirable collaboration partners to other groups. This theory is fascinating considering that humans are unique in their ability to foster tight bonds with other groups in the absence of any previously defined kinship (Pinker, 2010). Music and dance are also important aspects of visits and ceremonies, often preceding the initiation of alliances (Hagen & Bryant, 2003). Thus, music and dance may have served the purpose of enemy and alliance display in the ancestral past, especially as it can be employed at group level, regardless of language barriers, and without proximity (Merker, 2000).

It may be argued that music serves several benefits as an indirect route to contact between cultures. Social anthropologists and musicologists have posited this inkling for many years. Social anthropologist and ethnomusicologist John Blacking (1977) supported the idea of transcultural communication through music, in that every human belongs to a greater universal community, in which music may be used to bridge cultures and create a sense of siblinghood. Further, musicologist Nicholas Cook (1998) suggests that we use music both as means of gaining insight into other cultures and to negotiate cultural identity. These notions suggest that there are universal elements of music that can be appreciated and applied across cultures.

Musical universals have been a hotly debated topic within ethnomusicology dating back to Wilhelm Wundt (1911, cited in Campbell, 1997) who attempted to prove that all

peoples have monophonic singing and use intervals. Ethnomusicologist, Klaus Wachsmann (1971) contends that the near universal aspect of music lies in the relationship between sound and mind. Hence, the study of sound structure alone may not be nearly as fruitful as to study its relationship to the psychological perception of that sound. Thus, by focusing on both the sender and the receiver, communication can be studied, in which music may be used as a candidate to bridge different cultures. This may by some be perceived as a romanticized view, and empirical research on music as a means of bringing cultures together is scarce beyond case studies (Harwood et al., 2016). There are; nevertheless, studies indicating that music can reduce cultural biases.

2.4 Music’s Potential to Influence Cultural Biases Even though cultural biases tend to be relatively stable over time, studies have indicated that people’s cultural evaluations can be changed using exposure to music. Rodríguez-Bailón et al. (2009) exposed Spanish participants to Flamenco, a musical style that originated in Romani populations. Romani people are generally negatively perceived in Spain, but the study indicated that the exposure to Flamenco reduced this negative evaluation. Moreover, Vuoskoski et al. (2017) found that participants who listened to either an Indian or a West African popular song, prior to an implicit association test, manifested less out-group bias for the culture whose music they had listened to, compared to the other culture. Additionally, they found that participants exhibiting high trait empathy were more likely to have reduced out-group bias as a function of music listening. Taken together, music may be able to either reduce prejudice and/or foster some social bond between members of different cultures. It may be difficult to account for the exact mechanisms behind this prejudice-reducing tendency in music. Floating intentionality – coined by Cross and Morley (2009) - suggests that music provides the advantage of not being specific in its meaning, due to lack of semantic content, and can therefore reduce the potential for reactance to the semantic message in lyrics. Harwood (2017) adds that this aspect establishes music as a very honest form of communication, which may indicate that the communicator of music, the culture, can be regarded as an honest messenger. Although recent research has laid the groundwork for an exciting avenue for prejudice research through music, there are potential drawbacks to this notion. Harwood (2017) points out that music can also be used in intercultural contexts to incite dislike, racism, hatred and in some contexts promote intergroup violence. Sexually violent and aggressive rap music for

instance, can excite both negative attitudes towards women (Fischer & Greitemeyer, 2006; Kistler & Lee, 2009) and elicit negative stereotypes and attitudes towards African Americans (Johnson et al., 2000). In more extreme forms, music has been used by rallies of Nazism and other White Supremacy affiliations (Corte & Edwards, 2008). It is important to highlight; however, that songs used for hostile purposes often contain lyrics with explicit meaning. Hence, it is difficult to differentiate the effects of lyrics and music. Nevertheless, Lennings and Warburton (2011) indicated that music without lyrics can still elicit aggressiveness, although to a lesser degree than music with violent lyrics in conjunction. In addition, antisocial heavy metal lyrics are perceived as more harmful than the same lyrics used in country music (Ballard et al., 1999). These studies indicate both that there is such a thing as aggressive-sounding music and that stereotypical beliefs of aggression exist for various musical genres. Moreover, Maher et al. (2013) demonstrated through three studies that when Irish participants were exposed to music without lyrics that defied expectations, meaning that the music sounded unconventional, they were more willing to bet higher wagers on the Irish national team in a rugby match, allocate lower social services budgets to a minority group, and administer higher jail sentences to an outgroup member. It should be noted; however, that the presentation of music used in these studies was not directly related to the evaluation of the outgroup, it was rather used as a primer preceding questions about the outgroups. Taken together, music can produce both positive and negative cultural evaluations. Thus, in order to disentangle the effect of music on cultural evaluations, an investigation of cultural dimensions and musical perceptions are called for.

2.5 Cultural Evaluation Through Music The question of how music may inform positive and negative cultural evaluations has not previously been systematically investigated. Two aspects of this relationship need to be explored before such an examination can take place. First, an outline of basic dimensions of intergroup evaluation and how music may be able to communicate these dimensions. Second, an outline of musical parameters that can account for the communicative aspects of music. The following sections will discuss these two aspects in turn.

2.5.1 Dimensions Contributing to Cultural Biases In order to understand the ways in which musical exposure informs cultural evaluation, it is important to recognize dimensions of cultural evaluation that contribute to cultural biases. One theory that attempts to explain how cultural biases can originate is the integrated threat theory (Stephan et al., 2000), which posits that groups are devalued as a result of perceived threats from that group. This means that the groups can be perceived as threatening regardless of whether that threat actually exists or not. Another theory is the stereotype content model (Fiske et al., 2002). This model is inspired by an evolutionary perspective and proposes that perceived group stereotypes are formed along two distinct and independent dimensions. Humans are predisposed to assess strangers on their intent to harm or help them (warmth dimension) and second to evaluate the stranger’s ability to act on that intention (competence dimension). Going back to the signaling coalition theory explained initially (Hagen & Bryant, 2003), this theory accounts for a group’s motivation and ability to act collectively, displayed through coordinated music making. This is indeed relevant to the competence dimension, which posits that after perceiving a group’s intent, the group’s ability to act out that intent is evaluated. Further, according to the stereotype content model, the groups considered low on both warmth and competence are the most likely to be dehumanized. These people can evoke affective contempt and are associated with activation in brain areas related to disgust, namely the insula, as well as decreased brain activation in the medial prefrontal cortex, which is associated with thinking about humans (Harris & Fiske, 2006). An important part of dehumanization is the refusal to allow outgroups unique human emotions, which separates humans from robots and animals (Haslam, 2006). Taken together, cultural biases and prejudice represent people’s evaluation of cultures on various social and emotional dimensions. Just how music may alter cultural evaluations on such dimensions requires a more thorough investigation of the communicative aspects of music.

2.5.2 Communicative Aspects of Music Considering that perceived threat, warmth and competence may affect cultural evaluations, there should be aspects of music which can communicate intended social facets for the communication through music to be effective. Additionally, dehumanization is a result of viewing the outgroup as lacking in emotions. Music as a complex artistic expression, one

that separates humans from animals, may therefore impede tendencies to view outgroups as primitive, especially if music can communicate affective states transculturally.

2.5.2.1 Social Intentions and Functions. Music has the ability to convey both social intentions and social functions through acoustic cues. Aucouturier and Canonne (2017) asked expert musicians to communicate five types of non-musical social intentions: domineering, insolent, disdainful, conciliatory and caring. They found that both musically trained and non- musically trained participants recognized social intentions through social cognitive abilities that relied on acoustical cues of temporal and harmonic coordination that emerged between the musicians’ dynamic interaction. Further, Mehr et al. (2018) conducted a cross-cultural study and exposed participants from all over the world to song excerpts from 86 small-scale societies used for dance, healing, love, and lullaby. The musical materials used in this study were very well controlled, as they only contained singing and no musical instruments. Despite participants being unfamiliar with the cultures, results indicated reliable cross-cultural inferences about song function for dance and lullaby. This study indicates that there are certain acoustical social elements in music which can be recognized transculturally and may serve some underlying biological function. Taken together, music may communicate social intentions and functions across cultures.

2.5.2.2 Emotional Intensions. Music can also convey and elicit emotional responses, observable at both neural and behavioral levels (Huron & Margulis, 2010; Omar et al., 2011; Panksepp & Bernatzky, 2002), and an arguably significant part of the communicative aspects of music lies in its emotional expression (Bryant, 2013; Gabrielsson & Juslin, 2003). The recognition of basic emotions in music has been found across different cultures. Balkwill and Thompson (2004) found that Japanese listeners were sensitive to the intended emotions of joy, sadness and anger in Japanese, Western and Hindustani music, the latter culture’s music being unfamiliar to Japanese listeners. In addition, their data indicated that listeners depended on certain cues for recognition. Perceived melodically complex songs were associated with anger and sadness, slow tempo with sadness, and fast tempo and perceived melodically simple songs with joy. This study indicates that sensitivity to emotions conceived through acoustical cues may transcend cultural boundaries.

Further, Fritz et al. (2009) demonstrated a universal recognition of three basic emotions through music listening. They presented Western music and Mafa music to Western and Mafa listeners. The Mafa participants used in this study was a population residing in the Extreme North in the Mandara mountain range in Cameroon, who have presumably never been exposed to Western music before due to cultural isolation. They found that both groups recognized the basic emotions of sad, happy, and scared above chance level from Western music. Taken together, these studies indicate that emotional intention can be recognized through music across cultures, and that some cultures may rely on similar acoustical cues to elicit these reactions in the listener. It further supports the notion that basic emotions can be recognized cross-culturally through non-verbal channels.

2.6 Musical Parameters that may Shape Intergroup Biases

It seems reasonable to assume that music has the ability to change people’s perceptions of others, for better or worse, and that these perceptions may be formed by the communicated messages inherent in the music. These communicated messages, in turn, may be related to the performer’s expression and the listener’s recognition of several acoustical cues in music. Therefore, in order to understand how music can inform people’s cultural evaluations, it should be beneficial to explore parameters in music and investigate how they may contribute to musical and cultural evaluation.

2.6.1 Originality

There are a number of ways to infer originality in music. In this paper, originality in music is considered as a melody’s degree of novelty or unusualness. Simonton (1980a) created an algorithm based on tone-transition probabilities. He analyzed the first 6 notes of 5046 melodies produced by 10 famous composers, and measured the frequency of combinations of notes occurring, thereby arriving at note transition probabilities. The less probable a combination was, the more original it was considered to be. He found that the fame of a musical theme was positively linearly associated with the originality of the theme. A later analysis including more themes (15,618) and additional composers, revealed that when melodies became too unusual, the popularity tended to slightly decline, indicating curvilinearity (Simonton, 1980b). Thus, if one assumes that musical preference is linked to the evaluation of other cultures, it may also be assumed that the more original a song is (to a certain extent), the more positively the culture in which it stems from will be evaluated.

2.6.2 Intervals

Intervals in melodies have many parallels to intervals in speech. Ross et al. (2007) investigated whether humans’ preference for the 12-tone chromatic scale may be rooted in interval relationships found in speech. Indeed, they found that the frequency relationships between the first two formants of English vowel phones represent all possible intervals within the 12-tone chromatic scale. Further, sad speakers, compared to happy speakers, tend to use lower pitch registers and lower pitch fluctuations (Fairbanks and Pronovost, 1939). This relationship is inherent in minor key musical themes too, which is often associated with sadness and incorporates lower overall pitch fluctuation compared to major key themes (Huron, 2008). It has also been found that large melodic intervals are more difficult to predict and sound less coherent than smaller intervals when played in sequence (Huron, 2001; Marguils, 2005; Russo & Thompson, 2005). In addition, Quinto et al. (2013) found that expressed happiness in melodies was associated with higher average interval size in speech and melodies compared with expressed sadness and fear, whereas higher pitch variability in speech showed the opposite trend. Therefore, larger average interval sizes in music may be related to positive emotional expressions, whereas high pitch variability in speech may be an expression of negative emotions, perhaps evident in the lack of regulation control on the speaker’s side (Quinto et al., 2013). Thus, assuming that emotional connotations in music predict cultural evaluation, larger average interval sizes may also be related to positive perceptions in the cultural source of its production.

2.6.3 Modality

Modality refers to the major/minor distinction in Western music which are rooted in interval structures (Crowder, 1984). In major mode, there is a semitone interval between steps three and four, as opposed to a semitone interval between steps two and three in minor mode. There is clear evidence that modality is connected with emotional connotations, with major being perceived as happy, and minor being perceived as sad (Gagnon & Peretx, 2003; Hevner, 1935; Hunter et al., 2010; Kastner & Crowder, 1990; Webster & Weir, 2005). Neuroscientific research also highlights this distinction. Minor melodies, compared to major melodies, are usually related to heightened activation in left medial and frontal gyri and left parahippocampal gyrus (Green et al., 2008; Khalfa et al., 2005), whereas minor chords played independently, compared to major chords, are related to activation in retrosplenial cortex, brain stem, amygdala, and cerebellum (Pallesen et al., 2005).

Whether this distinction in affective connotations is a universal innate tendency is open to some discussion. Dalla Bella et al. (2001) found that this affective labeling is not evident in children until age of five. Nevertheless, Bowling et al. (2012) showed that the major-minor scale relationship in Western music is similar to south Indian music, which also corresponds to affective connotations. Further, this correspondence also seems to be evident in the vocalizations in the English and Tamil language, with excited, happy speech incorporating more major intervals, and negative, sad speech incorporating more minor intervals. It is important to note; however, that although the affective connotations links minor scales to sadness, sad music is not necessarily disliked or deemed unpleasant, as preferences depend on a number of factors, such as personality, mood and musical training (Eerola & Peltola, 2016; Ladinig & Schellenberg, 2012; Vuoskoski et al., 2012). Considering that major mode is associated with happiness, but not associated with aspects relating to preference, it is difficult to predict how modality will affect the evaluation of cultures. Thus, how modality informs cultural dimensions is subject to an exploratory investigation.

2.6.4 Tempo

One influential study on the perception of musical parameters dates back to 1937, when Hevner varied the tempo for the same musical excerpts and found that adjectives such as happy and exiting were chosen to describe fast tempo, whereas dreamy and serene were chosen to describe slow tempo. Moreover, Rigg (1940) examined emotional responses to variations in tempo and found that college students tended to rate the musical pieces as happier as tempo increased. This finding has been replicated in a number of studies (Balkwill & Thompson, 1999; Gabrielsson & Juslin, 1996; Gundlach, 1935; Juslin, 1997; Peretz et al., 1998; Webster & Weir, 2005). It has also been found that children as young as 5 years are able to distinguish fast and slow tempo on the basis of this happy-sad connotation (Dalla Bella et al., 2001). However, it has been noted that changes in tempo may be associated with changes in arousal rather than emotion, which can also affect performance on various cognitive tasks. Balch and Lewis (1996) demonstrated that variations in tempo affected performance on state-depended memory, and that variations in tempo were related to changes in arousal but not mood. In addition, Husain et al. (2002) showed that faster tempo was related to increases in arousal, which in turn enhanced performance on a spatial task. The effect of tempo on arousal was also confirmed by Holbrook and Anand (1990) who indicated that

positive affect experienced through music tended to follow from an increase in arousal level due to increased tempo. Several other studies indicate that alterations in tempo are associated with changes in physiological responses, such as heart rate variability (Van der Zwag et al., 2011), skin conductance level (Carpentier & Potter, 2007), respiration rate (Khalfa et a., 2008), blood pressure (Han et al., 2011) and frontal left hemisphere activity (Trochidis, & Bigand, 2013). In turn, increases in tempo tend to influence speed on physical behavioral tasks. In music listening tasks, increased tempo has been associated with increased speed in walking (Franěk et al., 2014), reading (Kallinen, 2002) and driving (Brodsky, 2001). Increased tempo is also associated to greater preference. Kellaris and Rice (1993) found that classical-style music in fast tempo (120 bpm) was rated less irritating, sad, and depressing than slower tempo (60 bpm). In school-aged children, there has also been found a preference for faster tempo in traditional jazz music, which also increased with age until college (LeBlanc et al., 1988). Additionally, a study tested the interaction effects of tempo preference and exercise intensity, showing a main effect of tempo preferred at 140 bpm compared to 80 bpm, and 120 bpm compared to 80 bpm. In addition, an interaction effect indicated that medium and fast tempo music were preferred at low to medium exercise intensity, whereas only fast tempo music was preferred at high intensity (Karageorghis et al., 2006). Thus, previous studies demonstrate that alterations in music tempo lead to altered arousal and affective judgement differences. Although increased tempo seems to predict higher preference, there is no agreement on what an ideal tempo is. A variety of preferences has been suggested ranging from 70 to 130 bpm (Fraisse, 1982; Iwanaga & Tsukamoto, 1998; Moelants, 2003; Rosenfeld, 1985) and the answer may possibly dependent on situational context, style of music, familiarity and age (Dobrota & Ercegovac, 2015; Jaret, 1982; Karageorghis et al., 2006; McAuley et al., 2006; Moelants, 2003). Considering that faster tempo is associated with both higher preference and happiness, one might assume that the cultural source of fast tempo songs will be evaluated more positively, again assuming that musical perception is associated with cultural evaluation.

2.6.5 Dissonance

The debate on why humans prefer some intervals over others is one of the oldest and most debated topics in auditory perception (Bowling & Purves, 2015). The intervals usually perceived as pleasant have been termed consonant, whereas intervals often perceived as

unpleasant have been termed dissonant. Intervals form the basis of melody and chords when played either sequentially or simultaneously. The concept of dissonance in music merits some attention for several reasons. Firstly, dissonance is associated with both mental and physical unpleasantness (e.g. Blood et al., 1999). Secondly, there are disagreements as to why humans prefer consonance (e.g. Cousineau et al., 2012). Thirdly, there is an ongoing debate on whether the preference for consonance has an inherently biological cause or whether it is a Western culturally constructed musical phenomenon (e.g. McDermott et al., 2016). This final point allows for a fascinating association to be made between dissonance and cultural evaluation, with reference to models on intergroup attitudes.

There is no clear agreement as to what a consonant interval comprises, and perceived pleasantness of intervals differ between cultures and across time (Rossing et al., 2002). Nevertheless, there is a general agreement that simple frequency interval ratios, such as a unison (1:1), octave (2:1) and perfect fifth (3:2), are considered pleasant to Western listeners, whereas complex frequency ratios, such as semitone (16:15) and tritone (45:32), are considered unpleasant to western listeners (Cazden, 1980; Guernsey, 1928; Van de Geer et al., 1962; Virtala & Tervaniemi, 2017). Consonance and dissonance have also been linked to a range of affective connotations. Consonance is associated with pleasure, agreeableness, peace and love, whereas dissonance is associated with painfulness, unacceptability, hostility and fear (Costa et al. 2000). There has also been found a correlation between unpleasantness and dissonance which is associated with different brain regions including the right parahippocampal cortex, the left amygdala and hippocampus (Blood et al., 1999; Koelsch et al., 2006). Dissonance has also been shown to trigger reflex-like responses in the brain stem and behavioural avoidance (see Juslin & Västfjäll, 2008). The preference for consonance seems to be even stronger for musically educated and experienced listeners. McDermott et al. (2010) found that both strength of harmonicity and strength of consonance correlated with years of musical training. Additionally, Dellacherie et al. (2011) found that experienced music listeners reported more negative affect in response to dissonant music in addition to stronger physiological responses. This indicates that musical experience may strengthen both the preference for consonance and the aversion to dissonance, perhaps due to more exposure and familiarity for consonant intervals through Western music.

There has been a variety of approaches and explanations for human preference for consonant intervals. The oldest dates back to Pythagoras (6th century BC), who noted that

preference for some intervals was related to the fact that those combinations exhibited fundamental frequency ratios that formed small integer ratios. Hemholtz (1863/1954) revised this notion and introduced the concept of roughness (also called beating), positing that complex frequency ratios includes similar interacting frequencies that gives rise to repeated constructive and destructive interference. Indeed, intervals with more complex ratios contains many frequencies that fall within the same bandwidth of cochlear filters, resulting in roughness and the perception of unpleasantness (Plomp & Levelt, 1965). Although this notion has been the dominant theory in the perception of dissonance, other studies have indicated that the perception of consonance persist even when controlling for roughness (Cousineau et al., 2012; McDermott et al., 2010). Therefore, another approach has been to focus on the vertical structure of single tones, noting that tones that produce a harmonic series include pitches which are integer multiples of the fundamental pitch. This harmonic series is therefore thought to correspond to tone combinations that are considered consonant (Bowling & Purves, 2015). In addition, a fourth approach, which is based on the harmonicity principle, stems from research on human vocalizations. Human vocalizations, which are based on learnt tonal pattern matchers, are naturally harmonic, and the more voice-like a tone-combination is, the more it should be appreciated. Therefore, Schwartz et al. (2003) collected thousands of voice segments from various databases and found that when the frequency ratios were averaged across segments, popular intervals used in Western music appeared as peaks in the distribution, which in turn reflected consonance preferences.

Another unresolved mystery in consonance research is whether the preference for some intervals is innate and universal. Evidence in favour of this proposal comes from studies on infants and cross-cultural research (Fritz et al., 2009; Masataka, 2006; Trainor et al., 1998). Nevertheless, some studies have challenged this view, indicating that infants’ preference for consonance does not depend on consonance preference but rather on familiarization (Plantinga & Trehub, 2012). In addition, the Tsimané people, who have experienced minimal exposure to Western music, show no preference for consonant over dissonant music intervals, even though they are able to discriminate such intervals (McDermott et al., 2016). In addition, the theory of an innate universal consonance preference does not explain why beating and harmonizing on dissonant intervals are desirable in a number of widely separated cultures, such as Middle Eastern, North Indian, and Bosnian musical cultures (Vassilakis, 2005; Jordania, 2006).

Thus, whether consonant music is truly a Western construct remains subject to debate. Nevertheless, there is little doubt that Western listeners prefer consonant intervals in favour of dissonant intervals. This preference may be applied to the evaluation of cultures too, in which the cultural source of consonant music will be more positively evaluated. Two notions are put forward in favour of this preference in relation to cultures attitudes; the ease of processing, and similarity to own cultural output.

2.6.5.1 Fluency. One aspect to consider when treating music as a communication tool between groups are fluency models. The fluency concept describes the ease of processing stimuli (Reber et al., 2004) and is dependent on many factors, such as visual and auditory clarity, as well as repeated exposure and familiarity, all of which influence perceivers’ judgements (Alter & Oppenheimer, 2009; Belke et al., 2010; Lev-Ari & Keysar, 2010; Winkielman, 2006). Recent studies have found that fluency processing positively predicts social evaluations, which has led to the belief that ease of processing may be used to guide prejudice too (Lick & Johnson, 2015). This seems especially salient under conditions of uncertainty about an outgroup member’s social identity (Lick & Johnson, 2013). For example, one study indicated that when participants experienced difficulties in classifying which constituent race a particular face belonged to, biracial face compositions were rated less attractive and evoked less positive affect than monoracial faces (Halberstadt & Winkielman, 2014). Claypool et al. (2012) indicated that when white participants were exposed to photographs of repeated or novel white and black target individuals, they were more likely to categorize the novel faces as outgroup members, regardless of race. The social evaluation effect has also been reported with regards to ease of verbal communication and semantics. Pearson (2011) found that participants engaging in interracial interactions on a computer screen were less likely to report an interest in future interactions with their partner if there was a 1-second delay in audiovisual feedback compared to no delay. Laham et al. (2012) also indicated that hypothetical targets with names easy to pronounce were rated as more likable than participants with names difficult to pronounce. Finally, Lev-Ari and Kaysar (2010) presented participants with trivia statements spoken in English with varying degrees of accents. Participants rated the accented speeches as less trustworthy due to difficulty in understanding the accented speech.

These studies indicate that people prefer what is familiar and easy to understand. Considering consonance as a Western preference rooted in the ease of processing through

repeated exposure, the cultural source of consonant music may be rated more favorably than cultural source of dissonant music, presuming that the ease of communication is positively applied to cultures as a whole.

2.6.5.2 Similarity. Another aspect to consider in relation to dissonance is the principle of similarity for social attraction. People prefer other people who are similar to them, both in terms of interests, personality and physical characteristics (Byrne et al., 1967; Klohen & Luo, 2003; Singh & Hoo, 2000; Stroebe et al., 1971; Ziegler & Golbeck, 2007). Moreover, this similarity-attraction relationship can be applied to other groups of people as well. One theory that could help explain this is the self-categorization theory (Turner et al., 1987), where the change in self-categorization from unique individuals to group members explains a person’s attitudes and behaviours. Social groups will be perceived as separate as long as the difference between groups are larger than differences within groups. Therefore, individuals who are perceived as more similar to oneself are more likely to be perceived as ingroup members, and in turn, more likely to be positively evaluated. Additionally, similarity may be perceived as less threatening, due to the absence of challenges to own cultural values (Stephan et al., 2000). Several studies support such theories. Henderson-King et al. (1997) found a positive relationship between similarity and negative outgroup attitudes, but only if the outgroup posed a threat to the ingroup. Struch and Schwartz (1989) found a correlation between Israeli perceived value dissimilarity and aggression towards an ultraorthodox Jewish outgroup. Moreover, Hensley and Duval (1976) conducted an experiment on perceived similarity and opinions with regard to specific issues. They found that as similarity of opinions between the participants and the other group increased, the liking for that group also increased.

To Western listeners, consonant music, compared to dissonant music, should thus sound more similar to Western music. Indeed, shared musical preference leads to more social attraction due to value similarity (Boer et al., 2011; Lonsdale & North, 2009). Thus, considering that similarity may predict positive intergroup relations, consonant music should predict positive cultural evaluations.

Nevertheless, it should be noted that similarity to own culture may not necessarily exert positive intergroup effects. From a social identity perspective (Tajfel & Turner, 1979), similarity between cultures may evoke negative feelings towards outgroups, as similarity may pose a threat to groups striving for distinctiveness. This perspective has been supported by several studies (Deschampes & Brown, 1983; Hornesey & Hogg, 2000; Jetten et al., 1996).

These conflicting findings have been theorized to be dependent on certain moderators, such as importance of dimension and ingroup identification (Costa-Lopes et al., 2012; Jetten et al., 2001; Roccas & Schwartz, 1993). Indeed, culture-specific musical preference is also linked to national identity through musical ethnocentrism, indicating that people prefer music of their own culture due to the music representing a form of national pride and identity (Boer et al., 2013).

3 The Present Research

The current study aimed, through two studies, to shed light on the relationship between musical parameters, musical perception and cultural evaluation. This is an important an understudied aspect of music, considering the potential for music to increase contact and foster bonds between groups. Further, there is evidence that music can communicate certain social and emotional intentions, but how this applies to how cultures are evaluated is unknown. Study 1 therefore attempted to compare musical and cultural evaluations of simple folk music melodies from different parts of the world to investigate whether certain subjective musical ratings could predict cultural evaluations. Additionally, this study investigated whether there were certain objective musical parameters, some of which have already previously been shown to affect evaluations of music, that could explain participants’ musical and cultural evaluations. Two of these parameters, general pitch variation and melodic velocity has, to the best of our knowledge, not been previously investigated in relation to affective connotations or preference.

Having obtained correlational insights in the first study, Study 2 set out to experimentally manipulate two notable musical parameters that have shown to predict emotional connotations and preference to music, namely tempo and dissonance. Additionally, this study investigated if musical training would interact with dissonance in the evaluation of cultures, since this participant group have been shown to report stronger preference for consonance and greater aversion to dissonance.

As already indicated, there are a variety of ways to investigate musical elements and how they relate to perception, such as lyrics, coalition quality, and dynamic coordination between performers. Nevertheless, such elements introduce a number of covariates, making inferences less trustworthy. Therefore, in order to avoid the potential influence of confounders and covariates, it was decided that MIDI-renditions of simple monophonic melodies would provide a fruitful strategy to investigate musical parameters.

In each study, we tested the effects of musical parameters on perceptions of basic dimensions of intergroup evaluation, namely warmth, threat, competence and evolvedness, the latter being a measure of dehumanization.

3.1 Hypothesis

Due to the exploratory nature of the study, overly specific predictions were kept to a minimum. The aim was to offer a general account of musical elements and how these inform participants of the cultural source of that music. Nevertheless, some predictions were tested. Firstly, it was expected that musical ratings would predict cultural ratings. The general prediction was that a more positive musical evaluation would be related to a more positive evaluation of the culture associated with it. Further, it was hypothesized that musical parameters would at least partly explain some of these effects. Considering that originality, interval size and modality are related to preference and positive emotional connotations in music, it was expected that these preferences and emotional connotations would be applied to the evaluations of cultures too. Moreover, considering the effect of tempo and dissonance on affective ratings, it was hypothesized that these musical parameters would have an effect on cultural evaluations. Specifically, faster tempo is associated with preference and happiness, and it was therefore expected that faster tempo would be associated with more positive cultural evaluations. Moreover, it was expected that dissonance, given its connections to unpleasantness and aversion, would exert a negative influence on evaluations of a culture’s warmth, competence and evolvedness, and positive associations with how threatening the culture is perceived to be. This relationship was predicted to be especially pronounced for people who have received musical training. Finally, considering that faster tempo is associated with increased arousal, it was expected that the effect of dissonance may interact with faster tempo, increasing perceptions of threat when dissonant, and increasing perceptions of warmth when consonant.

4 Study 1

The present study attempted to investigate whether objective musical parameters and subjective musical ratings could predict four dimensions of participants’ subjective cultural evaluations: warmth, threat, competence and evolvedness. Objective parameters and subjective ratings were obtained for short melodies of selected folk music presented through an online survey and analyzed using linear mixed model regression analyses. It was expected that subjective musical ratings would explain variance in cultural evaluations, at least on corresponding dimensions. Moreover, it was expected that musical parameters previously associated with preference and happiness, namely originality, interval size and modality, would be related to positive ratings of music and evaluations of cultures. No specific predictions were made with regards to general pitch variation and melodic velocity, as no relationship could be found between these musical parameters and preference or emotional connotations in the scientific literature.

4.1 Methods

4.1.1 Participants Participants were recruited via Amazon Mechanical Turk. The target sample size was set to approximately 50 participants. The sample size was chosen based on a rule of thumb by Hox (1998) for multilevel design with a minimum of 20 units at level 1(songs) and 50 units at level 2 (subjects). In addition, Arend and Schäfer (2019) conducted a large-scale simulation study based on input parameters most frequently analyzed by psychologists. They reported that a level 2 sample of 50 and a level 1 sample of 30, with high ICC would be able to detect a L1 minimum effect size of .15. However, differentiation between level 1 and level 2 effects was not applicable to the current study as all level 1 and level 2 samples were crossed (nested within each other). A total of 72 participants were initially recruited. Of these participants, 52 participants were selected (female=17, male = 35, age range = 25-72, Mage = 41.33, SDage=12.24) as they identified as White American. The selection was done in order to minimize cultural variation and prevent the possibility that some participants may belong to the foreign cultures of the music they were asked to rate.

4.1.2 Song Stimuli First, melodies were collected from various MIDI-databases: The Essen Folk Song Collection, the Finnish Folk Song database and the Densmore database. 30 songs were selected altogether: Seven songs were selected from North America (Native American), 13 from China, eight from Europe, and two from Africa. All songs were simple monophonic melodies. The number of selected songs from each continent was uneven due to difficulties in finding African folk songs of similar length and style. Next, to increase experimental control, each song was altered using GarageBand for Mac OS X. Tempo of each song was changed to 100 bpm and to a C (major or minor) scale so that the majority of notes for each song fell within the same pitch register (C3 to C4). In cases where the melody was inherently incomplete (stopped in the middle of a melodic phrase), the primary note was included in the end so that perception of incompleteness did not affect ratings. In addition, to control for timbre, all songs were converted to the sound of a flute instrument, selected from the GarageBand library. Each song lasted approximately 21 seconds. The melodies can be found at: https://osf.io/8wqpt/quickfiles

4.1.3 Procedure This study, together with Study 2, was approved by the University Faculty’s internal research Ethics Committee. After participants had accepted an invitation via Amazon Mechanical Turk, they were directed to a Qualtrics questionnaire. First, after receiving the informed consent form, participants were given instructions that they would listen to various songs of cultures that would be unknown to them. Subsequently, participants were asked to use headphones in order to minimize distraction from the surrounding environment. Next, participants listened to one song at a time, and then answered 7 VAS-scales ranging from 0 to 100 after each song. The ratings for music were: musical warmth – how warm/affectionate/kind does the music sound? (cold to warm), musical threat – how threatening does the music sound? (non-threatening to threatening), musical energy – how energetic is the music? (non-energetic to energetic), musical liking – how much do you like the music? (dislike to like), musical happiness – how happy does the music sound? (not happy at all to very happy), musical advancement – how advanced does the music sound? (primitive to advanced) and musical familiarity – how familiar does the music seem? (unfamiliar to familiar). Third, participants were presented with a question in a multiple-choice format asking about geographical region – where in the world do you think the melody is from? (Africa,

Oceania, East Asia, Middle East, Eastern Europe, Western Europe, North America, or South America). This question was presented to be used in the subsequent study for song selection in which a subset of songs would be used. The presentation order of the scales was randomized for each participant. Thereafter, participants were provided with a short text stating that the music excerpts they had listened to actually came from different, relatively unknown cultures. They were then given instructions that they would listen to all the excerpts one more time and rate the culture of which the music stemmed from on 6 VAS-scales after each song. The ratings for cultures were: cultural warmth – how warm/affectionate/kind does the culture seem (cold to warm), cultural competence – how competent does the culture seem? (primitive to advanced), cultural threat – how threatening does the culture seem (non-threatening to threatening), cultural liking – how much do you think you would like the culture? (dislike to like), cultural happiness – how happy does the culture seem? (not happy at all to very happy) and cultural evolvedness (ascent of man scale: Using the sliders below, indicate how evolved you consider the culture to be). The standardized Ascent of Man Scale (Kteily et al., 2015) is a validated blatant measure of dehumanization, which depicts five silhouettes illustrating five different stages of human evolution from apes to humans. People can use a slider bar to indicate which of the five stages they consider a human culture or group to be at. The presentation order of these scales was also here randomized for each participant. Rating scales and instructions are depicted in Appendix A. By first letting participants in random order rate all songs on dimensions of musical perception and then, again in random order, rate the cultures from which the music originated, participants were not able to match these ratings. Finally, at the end, participants completed a demographic questionnaire recording gender, age, musical training, and musical genre preference. The study lasted approximately 40 minutes and participants were awarded 7$ for participation.

4.1.4 Analyses The present study employed a crossed random effects design which is illustrated in Figure 1. Zero-order correlation analyses were first conducted to investigate the correlation between musical and cultural ratings. These ratings were then standardized (z-scored) in order to treat all variables (including parameters) on the same continuum. Additionally, a visual inspection of residuals indicated no violation of normality or homoscedasticity.

Figure 1

Crossed random Random effects Effects design Design of Study of1 Study 1

Subject 1 Subject 2 Subject 3 Subject 52

Song 1 Song 2 Song 3 Song 4 Song 30

Two sets of analyses were conducted. First, the musical parameters of the song excerpts were extracted using the MIDI-toolbox (Toiviainen & Eerola, 2016) and the MIR- toolbox (Lartillot & Toiviainen, 2007) for Matlab. All parameters were z-scored. The parameter descriptions can be found in Table 1. Second, linear mixed model regression analyses were conducted to create 15 models. These analyses were performed using the lme4 R-package (Bates et al., 2015). Linear mixed-effects models have the advantage of allowing for flexibility to specify both fixed and random effects. In our models with could therefore account for the unique effect between participants and songs. The first seven models investigated musical ratings on musical parameters by creating one model for each of the musical ratings (dependent variables). The second four models investigated cultural evaluation on musical parameters by creating one model for each of the cultural evaluations (dependent variables). Finally, the final four models investigated cultural evaluation on musical ratings by creating one model for each of the cultural evaluations (dependent variables). Participants and songs were entered as random intercepts. They were also entered as random slopes for respective within-participant and within-song predictors. P- values were obtained using the lmertest R-package (Kuznetsova et al., 2017), which uses Satterthwaite approximations for degrees of freedom. The model equations can be found in Appendix B and unstandardized regression coefficients of cultural evaluation on musical ratings and musical parameters can be found in Appendix C. Although not a standard procedure, all the predictors in each model were entered simultaneously and retained. There are two reasons for this. The first reason is that a stepwise method would require more specific ideas on the level of importance of each of the predictors, which due to the exploratory nature of the study, we did not have. The second reason is that we wanted to ensure that explained variance was attributed to the respective predictor. As musical parameters were between-song variables they were not considered dependent variables.

Table 1

Musical Parameter Descriptions

Musical parameters Descriptions

Originality This parameter refers to melodic complexity. This algorithm by Simonton (1984) is based on tone- transition probabilities. He analyzed the first 6 notes of 15,618 melodies, and measured the frequency of combinations of notes occurring, thereby arriving at note transition probabilities. The less probable a combination is, the more original it is considered to be. The formula is provided as a function in the MIDI-toolboxb

GPVa This parameter refers to the general number of pitch classes (tones) used to construct the melody. If a melodic phrase consists of the notes c-c-f, this equals to a GPV-value of 2. This is used as a measure of melodic complexity, as the fewer the number of pitch classes, the simpler the melody is considered to be.

Interval size This parameter is a measure of distance between successive intervals. It is a linear scaling of interval probabilities in which larger intervals are given more weight (larger values) than shorter ones. Each interval frequency percentage of the melody is multiplied by the number of semitones within that interval and then added together. If melodic intervals are visualized as successive ladder steps going up and down, melodies with longer successive steps (intervals), will be assigned higher values. If for example a major second is given the frequency value of 20 percent, it is given the value of 0.4: 2(semitones) X 0.2(major intervals). Interval distribution is provided as a function in the MIDI-toolboxb. Scaled function: q(I) = |I| * p(I), where p(I) refers to the probability of the interval occurring.

This parameter refers to the number of pitch onsets. The higher the number of pitch onsets, the shorter Melodic Velocity the interonset intervals are. The parameter is measure of ‘richness’ of the melody; the amount of notes used in a melody within a given number of bars. The formula is inspired by Christensen and Nettl’s (1960) conceptualization of melodic tempo: melodic velocity = no. of pitch onsets X 60 / duration of melody in seconds. It should be noted that melodic velocity is measured within a set number of bars and is therefore independent from tempo.

Modality This refers to estimation of the modality, i.e. major vs. minor, returned as a numerical value between -1 and +1: the closer it is to +1, the more major the given excerpt is predicted to be, the closer the value is to -1, the more minor the excerpt is. The formula is taken from the MIR-toolboxc

a GPV: General Pitch Variation. b Toiviainen and Eerola (2016) c Lartillot and Toiviainen (2007)

4.2 Results 4.2.1 Zero-order Correlations Unstandardized correlations, means and standard deviations between musical ratings and cultural ratings are displayed in Table 2. In general, correlations were high. However, neither standardized nor unstandardized model regressions violated assumption of multicollinearity VIF < 5.

Table 2 Unstandardized means, Means, standard Standard deviations Deviations and Pearson and zero-order Pearson correlation Zero-order matrix Correlation for continuous Coefficients variables for Continuous Variables

M SD 1 2 3 4 5 6 7 8 9 10 11

1. Musical Warmth 53.07 22.79 -

2. Musical Threat 23.29 22.68 -.52** -

3. Musical Liking 54.91 22.32 .56** -.35** -

4. Musical Happiness 47.22 47.22 .71** -.46** .50** -

5. Musical Energy 47.94 24.67 .58** -.33 .50 .72** -

6. Musical Advancement 50.49 22.14 .37* -.17 .46 .42* .49* -

7. Musical Familiarity 40.19 26.22 .30* -.18* .41* .26* .22* .25 -

8. Cultural Warmth 56.23 23.08 .46** -.37** .36** .48** .44** .30 .15* -

9. Cultural Threat 27.65 24.56 -.37** .36** -.27** -.32** -.28** -.18 -.04* -.52** -

10. Cultural Competence 63.18 23.06 .33* -.28** .36** .36* .37** .31* .25 .48** -.28** -

11. Cultural Evolvedness 80.65 21.76 .36** -.30** .38** .38** .37** .34* .04 .52** -.36** .69** -

Note. M = Mean. SD= Standard deviation. n=52. Responses ranged from 0-100. *p<.05. **p<.01. ***p<0.001

4.2.2 Linear Mixed Models 4.2.2.1 Effect of Objective Musical Parameters on Musical Ratings. The standardized regression estimates of subjective musical ratings on objective musical parameters can be found in Table 3. Melodic velocity was the most notable parameter effect, compared to the other musical parameters, and exerted the strongest effect on the musical ratings. Melodic velocity had a moderate, significantly positive effect musical warmth (β = .370, p < .001, d = 0.5), a weak, significantly negative effect on musical threat (β = -.192, p < .001, d = 0.2), a moderate, significantly positive effect on musical liking (β = .283, p < .001, d = 0.4), a moderate, significantly positive effect on musical happiness (β = .483, p < .001, d = 0.6), a strong, significantly positive effect on musical energy (β = .572, p < .001, d = 0.8), a moderate, significantly positive effect on musical advancement (β = .269, p < .001, d = 0.4), and a weak, significantly positive effect on musical familiarity (β = .131, p < .001, d = 0.2). The second most notable parameter, compared to other parameters in number of predicted musical ratings, was modality. Major melodies exerted a weak, significantly positive effect on musical warmth (β = .160, p < .001, d = 0.2), a weak, significantly negative effect on musical threat (β = -.151, p < .001, d = 0.2), a weak, significantly positive effect on musical happiness (β = .112, p < .001, d = 0.2), a very weak, significantly positive effect on energy (β = .051, p =.011, d = 0.1) and musical advancement (β = .043, p =.011, d = 0.1), and very weak, significantly negative effect on musical familiarity (β = -.047, p =.015, d = 0.1). Further, originality weakly, significantly positively predicted musical liking (β = .137, p < .001, d = 0.2), musical energy (β = .044, p =.014, d = 0.2), musical advancement (β = .099, p < .001, d = 0.1), and very

weakly, significantly positively predicted musical familiarity (β = .079, p =.002, d = 0.1). Intervals size very weakly, significantly positively predicted musical liking (β = .080, p= .023, d = 0.1), weakly, significantly positively predicted musical happiness (β = .104, p= .002, d = 0.2) and musical energy(β = .127, p < .001, d = 0.2), and very weakly, significantly positively predicted musical advancement (β = .075, p = .032, d = 0.1). Finally, general pitch variation exerted a very weak, significantly negative effect on musical happiness (β = -.056, p = .013, d = 0.1), and a very weak, significantly positive effect on musical advancement (β = .054, p = .020, d = 0.1).

Table 3 StandardizedRegression estimates Regression for musical Coefficients ratings on musical for Musicalparameters Ratings on Musical Parameters Warmth Threat Like Happiness Energy Advanced Familiarity Intercept = .000 Intercept = .000 Intercept = .000 Intercept = .000 Intercept = .000 Intercept = .000 Intercept = .000 SE = .052 SE = .078 SE = .074 SE = .049 SE = .044 SE = .076 SE =.098 Variables df = 51.04 df = 51.00 df = 51.00 df = 51.00 df = 51.00 df = 51.00 df = 51.00 t-value = .000 t-value = .000 t-value = .000 t-value = .000 t-value = .000 t-value = .000 t-value = .000 p = 1.000 p = .999 p = 1.000 p = .999 p = .999 p = 0.999 p = 0.999

Originality .045(.030) -.009(.009) .137(.029) .032(.028) .044(.027) .099(.028) .079(.026) p=0.125 p=0.748 p<0.001 p=0.255 p=0.014 p<0.001 p=0.002

GPV -.029(.549) .013(.023) .025(.023) -.056(.023) -.034(.022) .054(.023) .032(.021) p=0.235 p=0.587 p=0.288 p=0.013 p=0.116 p=0.020 p=0.127

Interval size .058(.036), -.063(.035) .080(.035) .104(.034) .127(.033) .075(.035) .015(.032) p=0.109 p=0.070 p=0.023 p=0.002 p<0.001 p=0.032 p=0.644

Melodic Velocity .370(.034) -.192(.033) .283(.033) .483(.032) .572(.031) .269(.033) .131(.030) p<0.001 p<0.001 p<0.001 p<0.001 p<0.001 p<0.001 p<0.001

Modality .160(.022) -.151(.021) .003(.021) .112(.021) .051(.020) .043(.021) -.047(.019) p<0.001 p<0.001 p=0.912 p<0.001 p=0.011 p=0.042 p=0.015

Note. GPV=General Pitch Variation. Standard Deviations in parentheses. Significance below .05 level in BOLD

4.2.2.2 Effect of Objective Musical Parameters on Culture Evaluation. The standardized regression estimates of subjective cultural evaluations on objective musical parameters can be found in Table 4. In concordance to the first model, melodic velocity was the strongest predictor of the cultural ratings. Melodic velocity exerted a moderate, significantly positive effect on cultural warmth (β = .375, p < .001, d = 0.5), a weak, significantly negative effect on cultural threat (β = -.121, p = .005, d = 0.2), a weak, significantly positive effect on cultural competence (β = .237, p < .001, d = 0.3) and cultural evolvedness (β = .164, p < .001, d = 0.3). Further, modality exerted a weak, significantly positive effect on cultural warmth (β = .157, p < .001, d = 0.2), a weak, significantly negative effect on cultural threat (β = -.149, p < .001, d = 0.2) and a very weak, significantly positive effect on cultural evolvedness (β = .036, p =.034, d =

0.1). Originality very weakly, significantly positively predicted cultural warmth (β = .060, p = .044, d = 0.1), weakly, significantly positively predicted cultural competence (β = .101, p < .001, d = 0.2) and very weakly, significantly positively predicted cultural evolvedness (β = .065, p = .005, d = 0.1). Interval size weakly, significantly positively predicted cultural warmth (β = .132, p < .001, d = 0.2), very weakly, significantly negatively predicted cultural threat (β = -.100, p = .005, d = 0.1) and very weakly, significantly positively predicted cultural competence (β = .070, p = .034, d = 0.1). Finally, general pitch variation very weakly, significantly negatively predicted cultural warmth (β = -.068, p = .005, d = 0.1).

Table 4

StandardizedRegression estimates Regression for culturalCoefficients ratings for onCultural musical Ratings parameters on Musical Parameters

Warmth Threat Competence Evolved Intercept = .000 Intercept = .000 Intercept: .000 Intercept: 80.65 SE = .053 SE = .077 SE = .081 SE = .105 Variables df = 51.00 df = 51.00 df: 51.00 df: 51.00 t-value = .000 t-value = .000 t-value = .000 t-value: .000 p = .999 p = .999 p = .999 p = .999

Originality .060(.029) .013(.029) .101(.027) .065(.023) p=0.044 p=0.657 p<0.001 p=0.005

GPV -.068(.024) .030(.024) .033(.022) .024(.019) p=0.005 p=0.209 p=0.138 p=0.187

Interval size .132(.036) -.100(.035) .070(.034) .037(.028) p<0.001 p=0.005 p=0.034 p=0.189

Melodic velocity .375(.034) -.120(.034) .237(.032) .164(.027) p<0.001 p<0.001 p<0.001 p<0.001

Modality .137(.022) -.149(.022) .030(.021) .036(.017) p<0.001 p<0.001 p=0.143 p=0.034

Note. GPV=General Pitch Variation. Standard Deviations in parentheses. Significance below .05 in BOLD

4.2.2.3 Effect of Musical Ratings on Culture Evaluation. The regression estimates for cultural evaluation on musical ratings can be found in Table 5. Musical warmth very weakly, significantly positively predicted cultural warmth (β = .104, p = .029, d = 0.1), and weakly, significantly negatively predicted cultural threat (β = -.126, p = .009, d = 0.2). Musical threat weakly, significantly negatively predicted cultural warmth (β = -.177, p < .001, d = 0.2), moderately, significantly positively predicted cultural threat (β = .291, p < .001, d = 0.4) and weakly, significantly negatively predicted cultural competence (β = -.122, p = .001, d = 0.2), and evolvedness (β = -.101, p = .001, d = 0.2). Musical liking weakly significantly positively predicted cultural competence (β = .139, p < .001, d = 0.2) and cultural evolvedness (β = .108, p < .001, d = 0.2). Musical happiness weakly, significantly positively predicted

cultural warmth (β = .148, p < .001, d = 0.2). Musical energy weakly, significantly positively predicted cultural warmth (β = .151, p < .001, d = 0.2), cultural competence (β = .168, p < .001, d = 0.2), and cultural evolvedness (β = .100, p < .001, d = 0.2). Finally, musical advancement very weakly, significantly positively predicted cultural warmth (β = .075, p < .001, d = 0.1), and weakly, significantly positively predicted cultural competence (β = .121, p < .001, d = 0.2), and cultural evolvedness (β = .072, p < .001, d = 0.2).

Table 5

StandardizedRegression estimates Regression for cultural Coefficients ratings for on Cultural musical Ratingsratings on Musical Ratings Warmth Threat Competence Evolved Intercept = .003 Intercept = .023 Intercept = -.036 Intercept: -.042 SE = .05 SE = .060 SE = .071 SE = .030 Variables df = 47.00 df = 43.43 df = 47.63 df = 50.12 t-value = .081 t-value = .385 t-value = -,506 t-value: -.383 p = .936 p = .702 p = .615 p = .704

Warmth .104(.046) -.126(.046) -.026(.035) -.016(.030) p=0.029 p=0.009 p=0.529 p=0.593

Threat -.177(.029) .291(.035) -.122(.035) -.101(.029) p<0.001 p<0.001 p=0.001 p=0.001

Like .058(.036) -.040(.036) .139(.034) .108(.029) p=0.102 p=0.277 p<0.001 p<0.001

Happiness .148(.043) .019(.039) .043(.038) .076(.042) p<0.001 p=0.619 p=0.260 p=0.078

Energy .151(.038) -.062(.041) .168(.042) .100(.036) p<0.001 p=0.132 p<0.001 p=0.008

Advanced .075(.032) -.026(.031) .121(.032) .072(.028) p=0.024 p=0.399 p<0.001 p=0.014

Familiarity -.018(.030) .028(.032) .069(.036) .012(.027) p=0.542 p=0.379 p=0.062 p=0.691

Note. Standard Deviations in parentheses. Significance below .05 in BOLD

4.2.2.4 Interaction of Complexity and Interval Size. As one previous study (Quinto et a., 2013) indicated that average interval size and pitch variability are associated with different emotional connotations in speech, it was suspected that interval size and general pitch variation might interact for their shared significant predictions for musical perception and cultural evaluation. Although the present research used different measurement techniques for these parameters than used by Quinto et al., they are similar conceptually. Interval size is associated with average size between intervals and general pitch variation is associated pitch variability. Therefore, an interaction term was added to the models they both predicted, namely musical happiness and cultural warmth. The first model indicated no interaction of complexity and

interval size on musical happiness (β = -.134, p =.0.526). The second model; however, did reveal a significant negative interaction between general pitch variation and interval size on cultural warmth (β = -.085, p = .001, d = 0.1) although this interaction was very weak.

4.3 Discussion The current discussion will evaluate the findings in light of previous models and theories presented in the introduction. The results of this study indicate that melodic velocity seems to have an important influence on musical and cultural evaluations. The previous focus by other researchers on musical parameters such as modality and tempo may have overlooked this element of melodies. Melodic velocity, although related to time, is conceptually different from tempo as tempo is held constant. This means that the increased presence of quavers and semiquavers in comparison to crotchets, within a melody, will increase melodic velocity within a set tempo and number of bars. Melodic velocity moderately positively predicted musical warmth, musical liking, musical happiness and musical advancement, and strongly positively predicted musical energy. Further, melodic velocity also moderately positively predicted cultural warmth, cultural competence and cultural evolvedness. There was also a negative association between melodic velocity and musical and cultural threat, but these associations were weaker. This musical parameter should be taken into consideration when investigating not only the cultural source of the music, but also the music in itself. Melodic velocity is the very increase of pitch onsets, which makes the melody the richer. This increase may be associated with increased arousal as found when increasing tempo (Balck & Lewis, 1996; Husain et al., 2002). Such increase in arousal may provide people with a pleasing experience associated with positivity and happiness (Holbrook & Anand, 1990), a connotation which in turn is applied to the cultural source of that music. Additionally, it may be that the increase of pitch onsets becomes less threatening as the conveyed musical message becomes more bountiful, making it easier to apprehend the communicated message. As Harwood (2017) points out, music is an honest form of communication, and the increased presence of tones within a given time, may incite greater trust in the receiver. Further, modality, also exerted some influence on musical and cultural evaluations. However, this parameter exerted generally weaker effects on both musical and cultural ratings, compared to musical velocity. Major melodies were weakly associated with musical and cultural warmth, and minor melodies were weakly associated musical and cultural threat.

Finally, major was also weakly associated with happiness. This finding complements previous findings on major’s association with positivity and happiness (Gagnon & Peretz, 2003; Hevner, 1935; Hunter et al., 2010; Kastner & Crowder, 1990; Webster & Weir, 2005). Additionally, it also supports the fact that although minor is associated with increased sadness, it is not necessarily less liked (Eerola & Peltola, 2016; Ladinig & Schellenberg, 2012; Vuoskoski et al., 2012). The current study extends previous findings and demonstrates that major intervals influence the perception of warmth in a culture and is not related to aspects of competence, a dimension of ability that is independent from the warmth dimension. The investigation of interval size can also be discussed in light of previous research. Quinto et al. (2013) reported that happiness, compared to sadness and fear, in speech and music was associated with higher average interval size. In the current study interval size was weakly but significantly associated with higher musical energy, increased musical happiness and increased cultural warmth. This may indicate that the emotional connotations of energy and happiness in larger intervals are applied to the evaluation of higher warmth in cultures. Additionally, Quinto et al. reported that for speech, happiness was also associated with lower pitch variability, indicating that a happy voice may be associated with the speaker’s ability to control pitch regularity. Similarly, the current data supports the indication that higher general pitch variation is associated with decreased happiness in music and decreased warmth of cultures, although these relationships were very weak. Interestingly, considering that higher average pitch size and lower pitch variability are both associated with happiness, it was postulated that interval size and general pitch variation may interact in affecting musical ratings and cultural evaluation. Such an interaction was found for cultural warmth, indicating that the when the positive effect of low general pitch variation on cultural warmth evaluation increased, the positive effect of interval size on cultural warmth evaluation decreased. This means that as long as the pitch variation is low, interval sizes are of less importance for the culture to be perceived as warm and vice versa. However, this interaction effect was very weak, and should not be treated as strong evidence. Moreover, higher originality was weakly associated with more liking and energy for the music, which is consistent with previous research on this parameter on composer popularity (Simonton, 1980a). Additionally, melodies were on average quite simple. This may explain why the more original songs were better liked in this study, considering that too much originality may lead to a decline in preference (Simonton, 1980b). Moreover, the current study found a weak relationship between higher originality and higher competence for the associated culture, which may indicate that among simple melodies, the more unusual a

melody is, the more competent the cultural source of that melody is considered to be. This aspect of originality in cultural perception therefore seems to be relevant for the dimension of ability that competence reflects. The results also indicate quite a consistent effect of musical ratings on cultural evaluation. The most notable of these predictors was the musical threat rating, which moderately positively predicted cultural threat. This indicates that the perception of perceived threat in music may be the strongest indication of threat by the associated culture. The results also indicate that the evaluation of cultural warmth is not predominantly dependent on the evaluation of music in the same domain, as both musical happiness and musical energy, compared to musical warmth, were stronger predictors of cultural warmth. Additionally, the effect of musical happiness on cultural warmth, and the effect of musical liking on cultural competence also further illustrate the independence of the cultural warmth and cultural competence dimensions, suggesting that the perceived warmth of a culture through music is more dependent on the happiness associated with music, whereas the perceived competence of the culture through music is more dependent on the liking and perceived advancement of the music of the associated culture. Finally, reviewing the cultural evaluation through the magnifying glass of the stereotype content model (Fiske et al., 2002), both in terms of objective musical parameters and subjective musical ratings, the current results indicate that the perceived cultural warmth dimension, the culture’s intentions, is dependent on the music’s modality and interval size, as well as subjectively perceived happiness of music. Seriatim, the subjective perception of musical happiness was positively influenced by the originality and interval size of music. On the other hand, cultural competence, the culture’s ability to perform intentions, seems more dependent on originality of the music and the subjective liking and perceived advancement of this music. Seriatim, the subjective liking and advancement of the music were positively influenced by the objective originality of music. Melodic velocity as well as the perceived energy of music predicted both the cultural warmth and the cultural competence dimension. In turn, melodic velocity predicted musical energy. It is important to note that many of the regressions between the objective and subjective features of music and cultures included weak effect sizes, which questions their meaningfulness. Nevertheless, even when entering all predictors simultaneously, the associations were highly significant and consistent, indicating that both objective and subjective features of music inform cultural evaluation to some degree. An aspect to keep in mind is that similar studies on musical ratings have often used a multiple-choice format with

few alternatives with one true social or emotional intention (see Aucouturier & Canonne, 2017; Fritz et al., 2009; Mehr et al. 2018). This may lead to an overestimation of response accuracy. The present study used a large number of response formats ranging from 0 to 100, and participants may have had different response styles to such scales. In addition, we do not know whether the different songs were actually intended to convey any of these dimensions. These aspects may in part explain the large variation in responses. Moreover, the current study only investigated the musical parameters already present in melodies without experimentally changing these to allow for direct comparisons, making it difficult to determine their true effect. Additionally, the participants rated the cultural source of melodies after having rated the melodies once, making the actual causal effect between musical parameters and cultural evaluation debatable. Thus, Study 2 aimed to address these questions.

5 Study 2 The second study set out to investigate the direct effect of musical parameters on cultural evaluation, by retaining full control of parameters via alteration. Tempo and dissonance were chosen for parameter investigation for two reasons. First, they have been extensively studied in relation to musical preference and emotional connotations. Second, these two parameters allowed for alteration without changing the melodic structure in terms of pitch variation, pitch number or pitch direction. A recent study also found that the alteration from consonance to dissonance in the dynamic interplay between two performers decreased listeners ratings of social intentions related to affiliation (Aucouturier & Canonne, 2017). Such a social evaluation of dissonance may therefore extend to the evaluation of cultures. A subset of folk songs from Study 1 were utilized. Next, these songs were presented through an online survey and responses were analyzed using linear mixed-effects modelling. It was expected that both consonant and fast tempo melodies would yield positive cultural evaluations. Moreover, considering the relationship between increased tempo and arousal, and the relationship between dissonance and unpleasantness, it was expected that an interaction between increased consonance and increased tempo would predict higher cultural warmth, whereas increased dissonance and higher tempo would predict higher cultural threat. Finally, it was expected that the positive effects of consonance and the negative effects dissonance would be more pronounced for people who had received prior musical training.

5.1 Methods 5.1.1 Participants Uniformly to Study 1, participants were collected via Amazon Mechanical Turk. The power rationale was also the same as in Study 1; however, since we used manipulated song versions in four alterations, we decided that approximately 200 participants would suffice, with approximately 50 participants for each alteration. An initial total of 270 participants were recruited. Of these participants, 212 (female=78, male = 134, age range = 20-74, Mage =

38.51, SDage = 11.12) white Americans were selected, again to ensure that participants had no cultural affiliation with the folk songs selected.

5.1.2 Song Stimuli Selection Songs were selected based on the highest standard deviation of geographic location ratings from the dataset of Study 1. This approach was chosen for selection as participants found these songs the most difficult to locate; hence, it would possibly reduce some of the

pre-existing bias that participants may have toward specific song location estimates and previous exposure to these folk songs. Standard deviations of geographical region ranged from 1.40 to 2.88. Eight melodies were selected with a cutoff standard deviation of 2.30. Two of these songs were discarded for modulation. First, the song with the highest standard deviation (2.88) only contained two pitches, making it impossible to create dissonant intervals without moving a pitch more than one semitone. Second, the melody with the third highest standard deviation (2.47) contained a melodic structure that made it unfeasible to create dissonant intervals without changing more than one pitch and hence the melodic structure. Therefore, six melodies were used as stimuli for this study. All melodies were manipulated to create dissonance and to alter tempo using GarageBand and Logic Pro X for Mac OS X. Each melody’s tempo was either increased 20 bpm or reduced 20 bpm so that each song was either presented in either 80 bpm or 120 bpm, whilst keeping pitch constant. Dissonance was induced according to a system, in which one tone inherent throughout the melody was changed either a semitone up or down thereby creating x number of dissonant intervals in the melody. The dissonant intervals created were either a major second changed to a semitone, or a perfect 4th or perfect 5th changed to a tritone (see Table 7). An example notation excerpt of the difference between a consonant and dissonant version of the same melody is illustrated in Figure 2.

Table 6 DissonanceDissonance manipulationsManipulations

Bars Key moved Dissonance in % of dissonant % of semiquavers SD of location C scale intervals created moved

Song 2 10 F to F♯ Tritone 17.9 20.6 2.31(8)

Song 3 9 D to D♭ Semitone 68.4 32.5 2.38(5)

Song 6 8 D to D♭ Semitone 35.7 31.3 2.36(6)

Song 14 9 F to F♯ Tritone 25.9 20.8 2.39(4)

Song 15 9 G to G♭ Tritone 21.1 31.6 2.31(7)

Song 29 8 F to F♯ Tritone 44.4 21.4 2.52(2)

Note. Percentage of semiquavers refers to the number of sixteenth note units that were moved to create dissonance. 1 moved quarter note (crotchet) equals 4 moved semiquavers. Ordering from highest to lowest standard deviation of location estimates in parentheses.

Prosjekt14_china_natmn Figure 2 =100 Flute Solo Difference Between a Consonant and a Dissonant Version of a Melody. .

Consonant1 : 14_diss_2 =100 Flute Solo . \ -# &Dissonant:\ ...... -# . 18 - . . . .

\ -# &Note. Excerpt. Q. from. . the first 6 .barsQ. of song. nr. 14. TheQ. .original. . (top). originated. . in. theQ. Han. dynasty. - #in 7China. (206 BC-220. AD)- . . -# .

5.1.3 Procedure & Q. Q. The. study,. together. with. Study. 1, was. approved. by the- #University Faculty’s internal research Ethics Committee. The procedure was similar to Study 1. Each participant listened to each of the six melodies once and rated the culture (based on the melody) on the six rating scales from the previous study. Importantly, for each song they were randomly assigned to listen to one of four possible variations: consonant-fast, consonant-slow, dissonant-fast or dissonant-slow. Participant instructions for Study 2 can be found in Appendix A. At the end, participants also completed a demographic questionnaire recording gender, age, musical training, and musical genre preference. The study took approximately 7 minutes to complete, and participants were awarded 1$ for participation.

5.1.4 Analyses The models created specified a crossed random effects design which is illustrated in Figure 3. In factorial terms, a 2 (tempo: slow versus fast) X 2(consonance versus dissonance) factorial design was utilized. Visual inspection of residuals revealed no violation of normality or homoscedasticity.

Figure 3

CrossedCrossed random Random effects Effects design Design of Study of2 Study 2

Subject 1 Subject 2 Subject 3 Subject 212

Song 1 Song 2 Song 3 Song 4 Song 6

Consonant/ Tempo Consonant/ Tempo Consonant/ Tempo Consonant/ Tempo Dissonant Dissonant Dissonant Dissonant

Consonant Dissonant Dissonant Consonant Fast Fast Slow Fast

Note. Random presentation of dissonance and tempo between songs (bottom level in figure).

All variables were z-scored so that all variables could be compared on equal continuums. Equivalently to Study 1, linear mixed model analyses were conducted to create 14 models. These analyses were performed using the lme4 R-package (Bates et al., 2012). Each categorical variable (musical parameters) was contrast coded, so that results could be interpreted in reference to the grand mean of all the data and not just the baseline of the other variable. The first four models aimed to investigate the effect of dissonance and tempo on the cultural ratings. These models also tested whether there was an interaction effect between the two parameters. In addition, due to the high number of musically trained participants in the sample, we explored group differences between people with musical training and people without musical training. Specifically, an additional six models aimed to investigate whether musical training could affect these cultural evaluations differently when controlling for other personal characteristics. Cultural liking and cultural happiness were added as dependent variables in order to examine a general trend in cultural ratings by musically trained participants. Moreover, the models examined whether musical training interacted with dissonance in predicting cultural evaluations. Finally, an additional four models were created in order to investigate whether the amount of dissonance present in the dissonant melodies could predict any of the cultural evaluations. Participants and songs were entered as random intercepts. They were also entered as random slopes for within-participant and within-song variables. P-values were obtained using the lmertest R-package (Kuznetsova et al., 2017) and boxplots were created for visualization using the ggplot2 R-package (Wickham, 2016). The

model equations can be found in Appendix B and unstandardized regression coefficients of cultural ratings on musical parameters can be found in Appendix C.

5.2 Study 2: Results 5.2.1 Linear Mixed Models 5.2.1.1 Effect of Dissonance and Tempo on Culture Evaluation. The standardized regression estimates of subjective cultural evaluations on dissonance and tempo can be found in Table 7. There was a strong, significantly negative effect of dissonance on cultural warmth (β = -.557, p < 0.001, d = 0.8) and a strong, significantly positive effect of dissonance on cultural threat (β = .453, p < 0.001, d = 0.7), a weak, significantly negative effect of dissonance on cultural competence (β = -.112, p = 0.012, d = 0.2) and cultural evolvedness (β = -.127, p = 0.003, d = 0.2). Participant ratings between consonance and dissonance for cultural evaluation are illustrated in Figure 4. Moreover, there was a weak, significantly positive effect of tempo on cultural competence (β = .137, p = 0.006, d=0.2) and a very weak significantly positive effect of tempo on cultural evolvedness (β = .080, p = 0.040, d = 0.1). Participant ratings between slow and fast tempo for cultural evaluation are illustrated are Figure 5. There were no interaction effects between dissonance and tempo on any of the cultural ratings.

Table 7 StandardizedRegression estimates Regression for Coefficientscultural ratings for Culturalon musical Ratings parameters on Musical Parameters

Warmth Threat Competence Evolved Intercept = -.008 Intercept = .000 Intercept = .000 Intercept = -.003 SE = .089 SE = .079 SE = .100 SE = .079 Variables df = 6.99 df = 14.68 df = 7.62 df = 16.03 t-value = -.090 t-value: -.001 t-value: -.007 t-value = -0.038 p = .931 p = .998 p = .995 p = .969

-.557(.084) .453(.099) -.112(.046) -.127(.042) Dissonance p<0.001 p=0.005 p=0.012 p=0.003

.190(.049) -.080(.052) .137(.049) .080(.038) Fast tempo p>0.05 p>0.05 p=0.006 p=0.040

.027(.091) .032(.078) .066(.090) .052(.075) Dissonance:Fast tempo p>0.05 p>0.05 p>0.05 p>0.05

Note. Categorical predictors are contrast coded. Standard Deviations in parentheses. Significance above 0.5 in BOLD

Figure 4 Boxplots of Mean Cultural Ratings by Subject between Consonance and Dissonance.

1 2

* * * * * 1 * 1 * *

0 0

−1 Threat Warmth Evolvedness Competence

0 −1 −1 Cultural threat Cultural Cultural warmth Cultural

Cultural evolvedness Cultural −2 Cultural competence Cultural

−2 −2 −1

Consonant Dissonant Consonant Dissonant Consonant Dissonant Dissonance Dissonance Consonant Dissonant Consonant Dissonant Consonant Dissonant Consonant Dissonance Dissonant Consonant Dissonance Dissonant

Note. Lines(whiskers) in boxplots indicate entire distribution of responses. Notch represents median. *p<.05. **p<.01. ***p<.001

Figure 5 Boxplots of Mean Cultural Ratings by Subject between Slow and Fast Tempo.

1 * * *

1 0

−1 Threat Warmth Evolvedness Competence

−1 0 −1

Cultural threat threat Cultural −2 Cultural warmth warmth Cultural Cultural evolvedness Cultural Cultural competence Cultural

−2 −2 −1

−3

Slow Fast Slow Fast Slow Fast Slow Fast Tempo Tempo Slow Fast Slow Fast Slow Tempo Fast Slow Tempo Fast

Note. Lines(whiskers) in boxplots indicate entire distribution of responses. Notch represents median. *p<.05. **p<.01. ***p<.001

5.2.1.2 Effect of Individual Differences on Culture Evaluation. The regression estimates for subjective cultural evaluations on musical training and other individual difference variables can be found in Table 8. There was a very strong, significantly positive effect of musical training on cultural threat after controlling for age and gender (β = .563, p = 0.001, d = 0.9). There was also a weak, significantly negative effect of age on cultural threat (β = -.157, p < 0.001, d = 0.2), and no interaction between the musical training and age on cultural threat. Moreover, there was a moderate, significantly positive effect of musical training on cultural competence (β = .314, p = 0.005, d = 0.4). As we suspected that the higher perception of threat and competence by people with musical training was due to these subjects generally experiencing greater affect in the evaluation of cultures through music, we tested whether people with musical training also rated the cultures as happier and more likable. Indeed, there was a moderate, significantly positive effect of musical training on cultural liking (β = .289, p = 0.009, d = 0.4) and cultural happiness (β = .222, p = 0.004, d = 0.4). Finally, we found a weak, significantly negative interaction effect between dissonance and musical training on cultural threat (β = -.195, p = 0.040, d = 0.3), denoting that the differential effects of consonance and dissonance on cultural threat ratings were smaller for people with musical training. This interaction is illustrated in Figure 6.

Table 8

StandardizedRegression estimates Regression for cultural Coefficients ratings on individual for Cultural differences Ratings on Individual Differences

Warmth Threat Competence Evolved Liking Happiness Intercept = -..053 Intercept = .081 Intercept = .073 Intercept = .060 Intercept = .058 Intercept = .039 SE = .077 SE = 079 SE = .096 SE = .079 SE = .091 SE = .079 Variables df = 7.99 df = 12.97 df = 8.67 df = 11.86 df = 4.42 df = 16.03 t-value = .687 t-value = 1.01 t-value = .767 t-value = .754 t-value = .637 t-value = -.038 p = .511 p = .328 p = .463 p = .465 p = .556 p = .708

-.127(.091) .112(.102) -.154(.103) -.242(.122) -.114(.102) -.071(.101) Gender p>0.05 p>0.05 p>0.05 p>0.05 p>0.05 p>0.05

Age -.001(.044) -.157(.049) .021(.048) .008(.059) -.054(.046) -.046(.047) p>0.05 p=0.001 p>0.05 p>0.05 p>0.05 p>0.05 .214(.107) .563(.103) .314(.108) .159(.121) .289(.107) .222(.122) Musical Training p>0.05 p<0.001 p=0.005 p>0.05 p=0.009 p=0.004

Musical Training:Dissonance .145(.113) -.195(.094) -.035(.095) -.071(.088) p>0.05 p=0.040 p>0.05 p>0.05

.014(.093) -.012(.104) -.061(.102) -.225(.118) Musical Training:Age p>0.05 p>0.05 p>0.05 p>0.05

Note. Categorical predictors are contrast coded. Standard Deviations in parentheses. Significance above 0.5 in BOLD

Figure 6 Boxplots of Mean Cultural Threat Ratings between Consonance and Dissonance, between Musically Trained and Non-musically Trained Participants.

MusicalMusTraining training

−0.5 Tz Non-trained Trained0.5

Cultural threat Cultural 0

−1

Consonant Dissonant Consonant Dissonant_con Dissonant

Note. Lines in boxplots indicate entire distribution of responses. Notch represents median. *p<.05. **p<.01. ***p<.001

5.2.1.3 Effect of Level of Dissonance on Culture Evaluation. The regression estimates for percentage of dissonant intervals and percentage of dissonant tones can be found in Table 9. This analysis was conducted using only the sample of dissonant melody versions (633 of 1272). We found no effect of amount of dissonance on any of the cultural ratings.

Table 9 StandardizedRegression estimates Regression for cultural Coefficients ratings for on Culturalamount of Ratings dissonance on Amount of dissonance Warmth Threat Competence Evolved Intercept = -.149 Intercept = -.074 Intercept = -.230 Intercept = -.105 SE = .309 SE = .269 SE = .256 SE = .211 Variables df = 3.00 df = 3.25 df: 2.02 df = 3.47 t-value = -.481 t-value: -.276 t-value: -.887 t-value = -.499 p = .663 p = .799 p = .468 p = .648

-.127(.162) .034(.141) -.193(.130) .120(.106) Percentage of dissonant intervals p>0.05 p>0.05 p>0.05 p>0.05

-.034(.285) .275(.285) .340(.267) .140(.215) Percentage of dissonant tones p>0.05 p>0.05 p>0.05 p>0.05

Note. Categorical predictors are contrast coded. Standard Deviations in parentheses.

5.3 Discussion The current discussion will evaluate the findings in light of previous models and theories presented in the introduction. Fast tempo predicted cultural competence and cultural evolvedness, indicating that participants perceived the cultural source of music at tempo 120 bpm as being more competent and evolved than at 80 bpm, although the effect of fast tempo on cultural evolvedness was very weak. The fact that tempo influenced perceived cultural competence and not cultural warmth or cultural threat may have to do competence’s affiliation with speed. As one develops competence in a skill, the speed at which the skill is executed will increase (Fernald et al., 1998; Taguchi, 2007). Moreover, faster tempo is by definition an increase in musical speed, which is also associated with increased bodily speed on behavioral tasks (Brodsky, 2001; Franěk et al., 2014; Kallinen, 2002). Therefore, there is a possibility that the elevation in musical tempo is associated with increased behavioral speed; hence, increased ability in other cultures. However, the existence of this speed-competence relationship in the cultural source of music is subject to further investigation. According to predictions, dissonance significantly predicted all the cultural ratings. Specifically, participants found the cultural source of consonant melodies warmer and the cultural source of dissonant melodies more threatening. The strong effect of this parameter further indicates that dissonance is an important factor in the evaluation of warmth and threat by a culture. These findings support Aucouturier and Canonne’s (2017) demonstration that increased manipulated dissonance leads to lower ratings of social affiliation. Additionally, participants found the cultural source of dissonance to be more culturally competent and evolved, although these relationships were weaker compared to the cultural warmth and threat dimensions. From a fluency perspective, consonance may be perceived as warmer and less threatening, due to consonance actually being easier to process, as a function of increased familiarity. Considering then music as a channel of communication, dissonant, unfamiliar music may be more difficult to process, due to the lack of previous exposure or understanding, which in turn makes the cultural source perceived more negatively. This is consistent with previous studies indicating that ease of communication and increased previous exposure elicit positive evaluations of an outgroup, especially when uncertainty exists about the outgroup’s social identity (Claypool et al., 2012; Lick & Johnson, 2013).

Further, the finding that consonant music positively predicted positive cultural ratings may indicate that western listeners prefer melodies that comprise consonant intervals, in other

words, intervals that are more commonly used in their own culture. This supports previous findings that cultures are more positively perceived when similarity to own culture increases (Henderson-King et al., 1997; Hensley & Duval., 1976; Struch & Schwartz, 1989).

It should also be noted that there is a possibility that participants rated dissonant melodies as more threatening due to evaluative conditioning, where a piece of music activates a particular response or emotion due to previous repeated pairing with an event (Juslin & Västfjäll, 2008). Bezdek and Gerrik (2008) appended that this music-memory association may extend to narrative content. They used the example of the famous movie theme soundtrack from Jaws (1975) conducted by John Williams, in which specific narrative content is foreshadowed by the dissonant minor second interval melody. Thus, it is possible that participants obtain specific associations to dissonant melodies, through movies and other media. Whether such an association can explain the evaluation of cultural sources of music may be considered in future studies.

Interpreting the current effects of dissonance and tempo on cultural evaluation with reference to the stereotype content model, the warmth and threat dimensions again display independence. Specifically, consonance increases the perception of warmth, the culture’s positive intent. Only on the competence dimension, the culture’s ability to carry out that intent, does tempo of the music exert an influence, with the importance of consonance decreasing. It should be noted, however, that tempo only exerted a weak effect on competence, and there may be other parameters in music that are important for the evaluation of this dimension in cultures.

There is also a possibility that the experienced unpleasantness followed by aversion and protective mechanisms that dissonance activates is responsible for the decreased perception of warmth and increased perception of threat in cultures. Nevertheless, this interpretation becomes more difficult to justify when reviewing results of cultural perception by participants with musical training. People with musical training experienced greater perception of threat, competence, liking and happiness for the cultures compared to individuals without musical training. It has previously been found that musically experienced individuals report grater activation and experience more arousal in response to sad and fear in music (Kreutz et al., 2008; Park et al., 2014). Musical expertise is also related to greater accuracy in recognizing emotions in music (Besson et al., 2007). This indicates that the greater threat evaluations by musically trained

participants can be attributed to these individuals recognizing more expressions of threat and competence in both consonant and dissonant melodies. Further, musical training negatively interacted with dissonance in predicting threat. This finding was somewhat surprising and countered with our predictions. Previous studies have shown that musically experienced individuals experience greater pleasantness to consonance and greater aversion to dissonance, indicating a greater span in perception between consonance and dissonance for musically experienced individuals (Dellacherie et al., 2011; McDermott et al., 2010). However, the current study indicated, with respect to cultures, that this span diminished when assessing the perceptions of threat in a culture. There may be a variety of ways to interpret this finding. First, it may be that although people with musical training experience greater unpleasantness and aversion to dissonance, they also have more experience with dissonant intervals, and therefore have a more objective view of consonance and dissonance when applying them to a cultural context, thereby decreasing the span in perception of cultural source of threat between consonance and dissonance. Second, from a social-categorization perspective (Turner et al., 1987), consonance, an indication of similarity to own cultures, should increase positive attitudes towards other cultures, unless moderators such as high ingroup identification or high dimension importance are present, in which case, perception of similarity may reduce this positive attitude (Costa-Lopez et al., 2012). Whether such moderating mechanisms, that could explain the reduction in consonance-dissonance effect for people with musical training, are present here remains only speculative and would require a more nuanced investigation of social identity in relation to music of other cultures. Moreover, there were no effects of amounts of dissonant intervals or amount of altered tones on any of the cultural evaluations. This could be due to participants creating a general impression of the culture by the mere presence of dissonance in the melody and are less concerned about, or not able to distinguish, the amount of dissonance throughout the melody. On the other hand, it could be that the percentage of dissonant tones within the scale did not differ enough (20.6-32.5 %) to exert any influence on cultural evaluation. Thus, perhaps with more variety in dissonance and longer song excerpts, the amount of dissonance could have contributed to differences in cultural evaluations. Finally, contrary to predictions, there were no interaction effects between tempo and dissonance on any of the cultural ratings. However, due to the short time length of songs, little variation in dissonance and only two tempo variations, we cannot affirm that such a relationship does not exist.

6. General Discussion

To the best of our knowledge, the two studies presented in this paper are the first to directly investigate whether listening to music from different cultures can influence how people evaluate these cultures on perceived warmth, threat, competence and evolvedness. It may also the first study to jointly investigate several musical parameters and their effect on cultural evaluation. Moreover, the results of the first study indicated that both musical and cultural evaluations are influenced by several musical parameters, most notably melodic velocity. This indicates that the increase in note onsets within a given tempo exerts a positive influence on both musical and cultural evaluations. This study also indicated that aspects of perceived elements in music, such as energy and advancement, positively predict positive cultural evaluations of warmth, competence and evolvedness, whereas negatively perceived elements of music, namely threat, negatively predicts the aforementioned cultural dimensions and positively predicts cultural threat. The second study aimed to test the effect of two musical parameters, namely dissonance and tempo, and found that dissonance strongly predicted warmth and threat, whereas both dissonance and tempo weakly predicted competence and evolvedness.

From the perspective of the stereotype content model, Study 1 indicated that the evaluation of cultural warmth is influenced by objective melodic velocity, modality and intervals of the music, and the perceived happiness and energy of the music. Study 2 appended that consonance strongly positively predicted cultural warmth. Moreover, Study 1 demonstrated that the evaluation of cultural competence, on the other hand, was influenced by objective melodic velocity and originality, and by liking and the perceived energy and advancement of the music. Study 2 appended that consonance and fast tempo weakly positively predicted cultural competence. Previous studies have shown that the combination of perceived warmth and competence may activate different behaviors and attitudes (see Cuddy et al., 2008). Thus, people may activate various reactions to different cultures through exposure to these musical parameters and ratings. However, the contextual aspects of music’s influence on cultural perception, such as when and where music will exert this effect, is subject to more investigation.

Musical training moderately positively predicted perceived cultural competence and very strongly predicted musical threat. Additional analyses also revealed that people with musical training evaluated the cultures as more likable and happier, which suggests that they

may experience or recognize stronger feelings towards cultures in general when evaluating these as a source of music. The results also revealed that older age predicts higher perceived threat, which signals that there may be other individual characteristics exerting effects on cultural evaluation. Further exploration of music and culture with regards to individual differences, such as age, needs to discern whether such individual characteristics will previse cultural evaluation through the presentation of music, or whether such effects are only inherent in the evaluation of cultures or the perception of music separately.

6.1 Limitations High experimental control was aimed to be established through a correlational study and a manipulation study. However, there are some limitations to these types of studies. First, there are many additional musical parameters the current studies did not assess. Only a flute instrument was used for the presentation of melodies. The timbre was hence held constant, but this timbre may still have influenced ratings. The current studies cannot account for such an influence. Second, the melodies were presented within a set pitch register. It would also indeed be interesting to investigate how perceptions of different pitch registers would influence cultural evaluation. Third, the study did not assess the influence of rhythm, beyond tempo and melodic velocity. Although each song were simple melodies presented without rhythmic instruments, there were still rhythmic cues present, such as pulse clarity, that might influence musical and cultural perception. Nevertheless, due to the large number of potential musical predictors to study, it was decided that the studies would focus more on tonal aspects. Fourth, there are many other parameters, related to dynamic aspects between performers, that can be investigated through music, such as interplay, harmony or number of performers. A study of these parameters; however, were considered beyond the scope of the present research, which only focused on simple melodies. Other parameters such as intonation and intonation contour may also exert important influences in evaluations; however, accounting for such cues would require an actual live performance and it does not make much sense to extract such parameters from MIDI files. These points considered above may also refer to the naturality of the study designs. The acoustical cues such as the timbre of the flute instrument, a set tempo of 100 bpm and a set pitch register may not have been very typical for the cultures the melodies stem from, where elements of timbre, rhythm, pitch and intonation may exert a great influence on social

expression and perception. Additionally, the type of music listening design that the current studies utilized, without providing any context to place the presented music in, may not represent true instances when music listening occurs. Thus, the drawbacks of the studies are that they may lack ecological validity in the presentation of music. The benefit of this approach, on the other hand, is that it allows for strong experimental control of variables.

Finally, the studies did not account for cultural biases already present in participants. The studies were an exploration of the effects of musical parameters and perceptions on cultural evaluations. Therefore, the current studies cannot make any assumptions with regards to the effect of music on pre-existing cultural biases. Additionally, the studies recruited white American participants, who all reported that they liked music, and results therefore cannot be applied to the broader population or to other cultures.

6.2 Future Directions The present research highlights interesting avenues for future research. Studies should investigate the effects of musical parameters further, by including other types of parameters, such as pitch register and timbre. It would also be interesting to explore whether the combination of certain musical parameters such as dissonance and tempo, that affects warmth and competence perceptions, will elicit different emotions or behaviors. The current study did not find any interaction between dissonance and tempo in predicting cultural evaluations. Therefore, the introduction of increased parameter variations in both dissonance and tempo, in addition to allowing participants to answer more detailed questions about the perceived culture, may reveal conjoint parameter effects in cultural evaluation and may also offer more insights into the exact nature of the effects of these parameters. Additionally, more nuanced investigations of perceived competence via faster tempo, may reveal the exact mechanisms behind the relationship between increased speed in music and higher competence in cultures. Moreover, future studies should explore just how musical and cultural evaluation interact with pre-existing intergroup bias, and whether certain musical parameters will enhance or reduce such biases. Empathic dispositions have also been shown to moderate the effect of music on cultural bias (Vuoskoski et al., 2017), and this could be taken into consideration in future research to investigate whether higher empathy interacts with musical parameters in predicting cultural evaluations. In addition, future studies could aim to recruit a less uniform group with participants from other parts of the world in addition to recruiting

people who do not care much for music, to unravel where and for whom music will exert an influence on cultural evaluations. Furthermore, people with musical training, in addition to finding the cultures more threatening based on their music, were also less affected by the difference in consonance and dissonance in their evaluation of the threat. This calls for an investigation of potential factors that could explain this finding, such as the importance of music as a dimension in evaluating cultural threat and perception of dissonance as an indicator of cultural threat. Considering that melodies have many linkages to speech (Juslin & Laukka, 2003), it would be interesting to investigate whether the musical parameters investigated in this study will have the same effect as speech parameters in predicting cultural evaluation. Considering that melodic velocity exerted a major influence on cultural evaluations, the mere presence of pitch onsets in speech may exert similar effects. It seems that musical parameters will indeed affect cultural evaluation in the absence of any other information about the culture. Future research should take this perspective into account and investigate to what degree these parameters are influential in isolation compared to the presence of other information. Music is rarely presented in isolation such as in the present study. Music is often presented in numerous contexts. Such contexts encompass a band being recommended by a friend or a song being played by a DJ at a social event, for example. In such instances the person introducing the piece of music may influence the perceived perception of the song and its cultural source. Music can also be enjoyed at concerts, where the performers and other audience members may influence perceptions of the music and its culture. Additionally, other elements within these contexts may also influence perceptions, such as the length of song and the number of performers. Just how such contexts interact with sound units will be of interest to studies examining just how much of an effect musical parameters and musical perception actually have in the presence of other information when evaluating the music of other cultures.

7 Conclusion This paper has attempted to shed light on the relationship between objective parameters in music, people’s subjective ratings of music and the evaluation of the cultural source of that music. In order to understand if music can effectively communicate social meanings, the elements within this communication need thorough and controlled investigations. The present research has offered a starting point in the evaluation of cultures through simple melodies. Hopefully, future research within the field of psychology and musicology can draw on this knowledge to put melodic elements in broader and perhaps more natural contexts in which contact may occur.

8 Funding The two studies presented in this thesis were funded by RITMO Centre for Interdisciplinary Studies in Rhythm, Time and Motion.

9 My Role in The Research Project The planning of the present research was done by my supervisors and me as a collaborative process. Based on our decisions, I collected and analyzed all data. I also performed the musical parameter extractions and manipulations. Finally, I wrote the thesis in full, with the help of pointers and comments by my supervisors.

References

Alter, A. L., & Oppenheimer, D. M. (2009). Uniting the tribes of fluency to form a metacognitive nation. Personality and social psychology review, 13(3), 219-235. https://doi-org.ezproxy.uio.no/10.1177/1088868309341564

Aucouturier, J. J., & Canonne, C. (2017). Musical friends and foes: The social cognition of affiliation and control in improvised interactions. Cognition, 161, 94-108. https://doi.org/10.1016/j.cognition.2017.01.019

Allport, G. W. (1954). The nature of prejudice. Perseus Books

Arend, M. G., & Schäfer, T. (2019). Statistical power in two-level models: A tutorial based on Monte Carlo simulation. Psychological methods, 24(1), 1. https://doi.org.ezproxy.uio.no/10.1037/met0000195

Bakagiannis, S., & Tarrant, M. (2006). Can music bring people together? Effects of shared musical preference on intergroup bias in adolescence. Scandinavian Journal of Psychology, 47(2), 129-136. https://doi-org.ezproxy.uio.no/10.1111/j.1467-9450.2006.00500.x

Balch, W. R., & Lewis, B. S. (1996). Music-dependent memory: The roles of tempo change and mood mediation. Journal of Experimental Psychology: Learning, Memory, and Cognition, 22(6), 1354. https://doi-org.ezproxy.uio.no/10.1037/0278-7393.22.6.1354

Balkwill, L. L., Thompson, W. F., & Matsunaga, R. I. E. (2004). Recognition of emotion in Japanese, Western, and Hindustani music by Japanese listeners 1. Japanese Psychological Research, 46(4), 337-349. https://doi.org/10.2307/40285811

Ballard, M. E., Dodson, A. R., & Bazzini, D. G. (1999). Genre of music and lyrical content: Expectation effects. The Journal of Genetic Psychology, 160(4), 476-487. https://doi.org.ezproxy.uio.no/10.1080/00221329909595560

Bates, D., Mächler, M., Bolker, B. & Walker, S. (2015). “Fitting Linear Mixed-Effects Models Using lme4.” Journal of Statistical Software, 67(1), 1–48. doi: 10.18637/jss.v067.i01.

Belke, B., Leder, H., Strobach, T., & Carbon, C. C. (2010). Cognitive fluency: High-level processing dynamics in art appreciation. Psychology of Aesthetics, Creativity, and the Arts, 4(4), 214. https://doi-org.ezproxy.uio.no/10.1037/a0019648

Besson, M., Schön, D., Moreno, S., Santos, A., & Magne, C. (2007). Influence of musical expertise and musical training on pitch processing in music and language. Restorative neurology and neuroscience, 25(3-4), 399-410. https://content-iospress- com.ezproxy.uio.no/articles/restorative-neurology-and-neuroscience/rnn253423

Bezdek, M. A., & Gerrig, R. J. (2008). Musical emotions in the context of narrative film. Behavioral and Brain Sciences, 31(5), 578-578. https://doi-org.ezproxy.uio.no/10.1017/S0140525X08005323

Binder, J., Zagefka, H., Brown, R., Funke, F., Kessler, T., Mummendey, A., ... & Leyens, J. P. (2009). Does contact reduce prejudice or does prejudice reduce contact? A longitudinal test of the contact hypothesis among majority and minority groups in three European countries. Journal of personality and social psychology, 96(4), 843. https://doi-org.ezproxy.uio.no/10.1037/a0013470

Blacking, J. (1977). Can musical universals be heard?. The World of Music, 19(1/2), 14-22. https://www-jstor-org.ezproxy.uio.no/stable/43560438

Blood, A. J., Zatorre, R. J., Bermudez, P., & Evans, A. C. (1999). Emotional responses to pleasant and unpleasant music correlate with activity in paralimbic brain regions. Nature neuroscience, 2(4), 382-387. https://doi.org.ezproxy.uio.no/10.1038/7299

Boer, D., Fischer, R., Strack, M., Bond, M. H., Lo, E., & Lam, J. (2011). How shared preferences in music create bonds between people: Values as the missing link. Personality and Social Psychology Bulletin, 37(9), 1159-1171. https://doi.org.ezproxy.uio.no/10.1177/0146167211407521

Boer, D., Fischer, R., Gonzalez Atilano, M. L., de Garay Hernández, J., Moreno Garcia, L. I., Mendoza, S., ... & Lo, E. (2013). Music, identity, and musical ethnocentrism of young people in six A sian, L atin A merican, and W estern cultures. Journal of Applied Social Psychology, 43(12), 2360-2376. https://doi-org.ezproxy.uio.no/10.1111/jasp.12185

Bowling, D. L., & Purves, D. (2015). A biological rationale for musical consonance. Proceedings of the National Academy of Sciences, 112(36), 11155-11160. https://doi-org.ezproxy.uio.no/10.1073/pnas.1505768112

Bowling, D. L., Sundararajan, J., Han, S. E., & Purves, D. (2012). Expression of emotion in Eastern and Western music mirrors vocalization. PLoS One, 7(3). doi: 10.1371/journal.pone.0031942

Brodsky, W. (2001). The effects of music tempo on simulated driving performance and vehicular control. Transportation research part F: traffic psychology and behaviour, 4(4), 219-241. https://doi.org/10.1016/S1369-8478(01)00025-0

Bryant, G. A. (2013). Animal signals and emotion in music: Coordinating affect across groups. Frontiers in Psychology, 4, 990. https://doi.org/10.3389/fpsyg.2013.00990

Byrne, D., Griffitt, W., & Stefaniak, D. (1967). Attraction and similarity of personality characteristics. Journal of Personality and social Psychology, 5(1), 82. https://doi.org.ezproxy.uio.no/10.1037/h0021198

Cameron, L., Rutland, A., Brown, R., & Douch, R. (2006). Changing children's intergroup attitudes toward refugees: Testing different models of extended contact. Child development, 77(5), 1208-1219. https://doi.org/10.1111/j.1467-8624.2006.00929.x

Campbell, P. S. (1997). Music, the universal language: Fact or fallacy?. International Journal of Music Education, (1), 32-39. https://doi.org.ezproxy.uio.no/10.1177/025576149702900105

Carpentier, F. R. & Potter, R. F. (2007). Effects of music on physiological arousal: Explorations into tempo and genre. Media Psychology, 10(3), 339-363. https://doi.org.ezproxy.uio.no/10.1080/15213260701533045

Cazden, N. (1980). The definition of consonance and dissonance. International Review of the Aesthetics and Sociology of Music, 123-168. https://www-jstor- org.ezproxy.uio.no/stable/836494

Claypool, H. M., Housley, M. K., Hugenberg, K., Bernstein, M. J., & Mackie, D. M. (2012). Easing in: Fluent processing brings others into the ingroup. Group Processes & Intergroup Relations, 15(4), 441-455. https://doi- org.ezproxy.uio.no/10.1177/1368430212439115

Cook N. (1998). Music. A very short introduction. Oxford University Press.

Corte, U. & Edwards, B. (2008). White Power music and the mobilization of racist social movements. Music and Arts in Action, 1(1), 4-20. https//:musicandartsinaction.net/index.php/main/article/view/white power music

Costa, M., Ricci Bitti, P. E., & Bonfiglioli, L. (2000). Psychological connotations of harmonic musical intervals. Psychology of Music, 28(1), 4-22. https://doi- org.ezproxy.uio.no/10.1177/0305735600281002

Costa-Lopes, R., Vala, J., & Judd, C. M. (2012). Similarity and dissimilarity in intergroup relations: Different dimensions, different processes. Revue internationale de psychologie sociale, 25(1), 31-65. https://www-cairn-info.ezproxy.uio.no/revue- internationale-de-psychologie-sociale-2012-1-page-31.htm?contenu=article

Cousineau, M., McDermott, J. H., & Peretz, I. (2012). The basis of musical consonance as revealed by congenital amusia. Proceedings of the National Academy of Sciences, 109(48), 19858-19863. https://doi.org.ezproxy.uio.no/10.1073/pnas.1207989109

Cross, I., & Morley, I. (2009). The evolution of music: Theories, definitions and the nature of the evidence. In S. Malloch and C. Trevarthen (Eds.), Communicative musicality: Exploring the basis of human companionship (pp. 61-81). Oxford University Press.

Crowder, R. G. (1984). Perception of the major/minor distinction: I. Historical and theoretical foundations. Psychomusicology: a journal of research in music cognition, 4(1-2), 3. https://doi-org.ezproxy.uio.no/10.1037/h0094207

Cuddy, A. J., Fiske, S. T., & Glick, P. (2008). Warmth and competence as universal dimensions of social perception: The stereotype content model and the BIAS map. Advances in experimental social psychology, 40, 61-149. https://doi.org/10.1016/S0065-2601(07)00002-0

Dalla Bella, S., Peretz, I., Rousseau, L., & Gosselin, N. (2001). A developmental study of the affective value of tempo and mode in music. Cognition, 80(3), B1-B10. https://doi.org/10.1016/S0010-0277(00)00136-0

Dellacherie, D., Roy, M., Hugueville, L., Peretz, I., & Samson, S. (2011). The effect of musical experience on emotional self‐reports and psychophysiological responses to dissonance. Psychophysiology, 48(3), 337-349. https://doi-org.ezproxy.uio.no/10.1111/j.1469-8986.2010.01075.x

Deschamps, J. C., & Brown, R. (1983). Superordinate goals and intergroup conflict. British Journal of Social Psychology, 22(3), 189-195. https://doi-org.ezproxy.uio.no/10.1111/j.2044-8309.1983.tb00583.x

Dobrota, S., & Reić Ercegovac, I. (2015). The relationship between music preferences of different mode and tempo and personality traits–implications for music pedagogy. Music Education Research, 17(2), 234-247. https://doi.org.ezproxy.uio.no/10.1080/14613808.2014.933790

Eerola, T., & Peltola, H. R. (2016). Memorable experiences with sad music—reasons, reactions and mechanisms of three types of experiences. PLoS One, 11(6). doi: 10.1371/journal.pone.0157444

Efferson, C., Lalive, R., & Fehr, E. (2008). The coevolution of cultural groups and ingroup favoritism. Science, 321(5897), 1844-1849. doi: 10.1126/science.1155805

Fairbanks, G., & Pronovost, W. (1939). An experimental study of the pitch characteristics of the voice during the expression of emotion. Communications Monographs, 6(1), 87- 104. https://doi.org/10.1080/03637753909374863

Franěk, M., van Noorden, L., & Režný, L. (2014). Tempo and walking speed with music in the urban context. Frontiers in psychology, 5, 1361. https://doi.org/10.3389/fpsyg.2014.01361

Fernald, A., Pinto, J. P., Swingley, D., Weinbergy, A., & McRoberts, G. W. (1998). Rapid gains in speed of verbal processing by infants in the 2nd year. Psychological Science, 9(3), 228-231. https://doi-org.ezproxy.uio.no/10.1111/1467-9280.00044

Fischer, P., & Greitemeyer, T. (2006). Music and aggression: The impact of sexual-aggressive song lyrics on aggression-related thoughts, emotions, and behavior toward the same and the opposite sex. Personality and Social Psychology Bulletin, 32(9), 1165-1176. https://doi-org.ezproxy.uio.no/10.1177/0146167206288670

Fiske, S. T., Cuddy, A. J., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype content: competence and warmth respectively follow from perceived status and competition. Journal of personality and social psychology, 82(6), 878. https://doi-org.ezproxy.uio.no/10.1037/0022-3514.82.6.878

Fraisse, P. (1982). Rhythm and tempo. In D. Deutsch (ed), Tbepsychology ofmusic(pp. 149- 80). Academic Press.

Fritz, T., Jentschke, S., Gosselin, N., Sammler, D., Peretz, I., Turner, R., ... & Koelsch, S. (2009). Universal recognition of three basic emotions in music. Current biology, 19(7), 573-576. https://doi.org/10.1016/j.cub.2009.02.058

Gabrielsson, A., & Juslin, P. N. (1996). Emotional expression in music performance: Between the performer's intention and the listener's experience. Psychology of music, 24(1), 68- 91. https://doi-org.ezproxy.uio.no/10.1177/0305735696241007

Gabrielsson, A. & Juslin, P. N. (2003). Emotional expression in music. Oxford University Press.

Gagnon, L. & Peretz, I. (2003). Mode and tempo relative contributions to “happy-sad” judgements in equitone melodies. Cognition and emotion, 17(1), 25-40. https://doi.org.ezproxy.uio.no/10.1080/02699930302279

Gilboa, A. & Bodner, E. (2009). What are your thoughts when the national anthem is playing? An empirical exploration. Psychology of Music, 37(4), 459-484. https://doi-org.ezproxy.uio.no/10.1177/0305735608097249

Green, A. C., Bærentsen, K. B., Stødkilde-Jørgensen, H., Wallentin, M., Roepstorff, A., & Vuust, P. (2008). Music in minor activates limbic structures: a relationship with dissonance?. Neuroreport, 19(7), 711-715. doi: 10.1097/WNR.0b013e3282fd0dd8

Guernsey, M. (1928). The role of consonance and dissonance in music. The American Journal of Psychology, 173-204. https://www-jstor-org.ezproxy.uio.no/stable/1414484

Gundlach, R. H. (1935). Factors determining the characterization of musical phrases. The American Journal of Psychology, 47(4), 624-643. https://www-jstor-org.ezproxy.uio.no/stable/1416007

Halberstadt, J. & Winkielman, P. (2014). Easy on the eyes, or hard to categorize: Classification difficulty decreases the appeal of facial blends. Journal of Experimental Social Psychology, 50, 175-183. https://doi.org/10.1016/j.jesp.2013.08.004

Hagen, E. H. & Bryant, G. A. (2003). Music and dance as a coalition signaling system. Human nature, 14(1), 21-51. https://doi-org.ezproxy.uio.no/10.1007/s12110-003-1015-z

Han, I., Kleifgen, E., Martin, C., & Zarling, E. (2011). The Physiological Response to Music Tempo, The Investigation of the'Pump-Up'Song. Journal of Advanced Student Science, 01, 435. http://jass.neuro.wisc.edu/2011/01/

Harris, L. T. & Fiske, S. T. (2006). Dehumanizing the lowest of the low: Neuroimaging responses to extreme out-groups. Psychological science, 17(10), 847-853. https://doi-org.ezproxy.uio.no/10.1111/j.1467-9280.2006.01793.x

Harwood, J. (2017). Music and intergroup relations: Exacerbating conflict and building harmony through music. Review of Communication Research, 5, 1-34. https://doi.org/10.12840/issn.2255-4165.2017.05.01.012

Harwood, J., Qadar, F., & Chen, C. Y. (2016). Harmonious contact: Stories about intergroup musical collaboration improve intergroup attitudes. Journal of Communication, 66(6), 937-959. https://doi.org/10.1111/jcom.12261

Haslam, N. (2006). Dehumanization: An integrative review. Personality and social psychology review, 10(3), 252-264. https://doi-org.ezproxy.uio.no/10.1207/s15327957pspr1003_4

Helmholtz, H.V. (1954) On the Sensations of Tone as a Physiological Basis for the Theory of Music (A.J. Ellis, Trans.). Dover Publications. (Original work published 1863).

Henderson-King, E., Henderson-King, D., Zhermer, N., Posokhova, S., & Chiker, V. (1997). In-group favoritism and perceived similarity: A look at Russians' perceptions in the post-Soviet era. Personality and Social Psychology Bulletin, 23(10), 1013-1021. https://doi-org.ezproxy.uio.no/10.1177/01461672972310002

Hensley, V., & Duval, S. (1976). Some perceptual determinants of perceived similarity, liking, and correctness. Journal of Personality and Social Psychology, 34(2), 159. https://doi-org.ezproxy.uio.no/10.1037/0022-3514.34.2.159

Hevner, K. (1935). The affective character of the major and minor modes in music. The American Journal of Psychology, 47(1), 103-118. https://www-jstor-org.ezproxy.uio.no/stable/1416710

Hevner, K. (1937). The affective value of pitch and tempo in music. The American Journal of Psychology, 49(4), 621-630. https://www-jstor-org.ezproxy.uio.no/stable/1416385

Hox, J. (1998). Multilevel modeling: When and why. In Classification, data analysis, and data highways (pp. 147-154). Springer. https://doi-org.ezproxy.uio.no/10.1007/978-3-642-72087-1_17

Holbrook, M. B. & Anand, P. (1990). Effects of tempo and situational arousal on the listener's perceptual and affective responses to music. Psychology of Music, 18(2), 150-162. https://doi-org.ezproxy.uio.no/10.1177/0305735690182004

Hornsey, M. J. & Hogg, M. A. (2000). Subgroup relations: A comparison of mutual intergroup differentiation and common ingroup identity models of prejudice reduction. Personality and Social Psychology Bulletin, 26(2), 242-256. https://doi-org.ezproxy.uio.no/10.1177/0146167200264010

Hunter, P. G., Schellenberg, E. G., & Schimmack, U. (2010). Feelings and perceptions of happiness and sadness induced by music: similarities, differences, and mixed emotions. Psychology of Aesthetics, Creativity, and the Arts, 4(1), 47. https://doi-org.ezproxy.uio.no/10.1037/a0016873

Huron, D. (2001). Is music an evolutionary adaptation?. Annals of the New York Academy of sciences, 930(1), 43-61. https://doi: 10.1111/j.1749-6632.2001.tb05724.x.

Huron, D. & Margulis, E. H. (2010). Musical expectancy and thrills. In P. N. Juslin & J. A. Sloboda (Eds.), Series in affective science. Handbook of music and emotion: Theory, research, applications (p. 575–604). Oxford University Press. https://psycnet-apa-org.ezproxy.uio.no/record/2010-02543-021

Husain, G., Thompson, W. F., & Schellenberg, E. G. (2002). Effects of musical tempo and mode on arousal, mood, and spatial abilities. Music Perception: An Interdisciplinary Journal, 20(2), 151-171. https://doi.org/10.1525/mp.2002.20.2.151

Husnu, S. & Crisp, R. J. (2010). Elaboration enhances the imagined contact effect. Journal of Experimental Social Psychology, 46(6), 943-950. https://doi.org/10.1016/j.jesp.2010.05.014

Iwanaga, M. & Tsukamoto, M. (1998). Preference for musical tempo involving systematic variations of presented tempi for known and unknown musical excerpts. Perceptual and motor skills, 86(1), 31-41. https://doi-org.ezproxy.uio.no/10.2466/pms.1998.86.1.31

Jaret, C. (1982). Hits or just heartaches: Characteristics of successful and unsuccessful country music songs. Popular Music & Society, 8(2), 113-124. https://doi.org.ezproxy.uio.no/10.1080/03007768208591184

Jetten, J., Spears, R., & Manstead, A. S. (1996). Intergroup norms and intergroup discrimination: distinctive self-categorization and social identity effects. Journal of personality and social psychology, 71(6), 1222. https://doi-org.ezproxy.uio.no/10.1037/0022-3514.71.6.1222

Jetten, J., Spears, R., & Manstead, A. S. (2001). Similarity as a source of differentiation: The role of group identification. European Journal of Social Psychology, 31(6), 621-640. https://doi-org.ezproxy.uio.no/10.1002/ejsp.72

Johnson, J. D., Trawalter, S., & Dovidio, J. F. (2000). Converging interracial consequences of exposure to violent rap music on stereotypical attributions of blacks. Journal of Experimental Social Psychology, 36(3), 233-251. https://doi.org/10.1006/jesp.1999.1404

Jordania, J. (2006). Who asked the first question? The origins of human choral singing, intelligence, language and speech. Tblisi, Georgia: Logos.

Juslin, P. N. (1997). Perceived emotional expression in synthesized performances of a short melody: Capturing the listener's judgment policy. Musicae scientiae, 1(2), 225-256. https://doi-org.ezproxy.uio.no/10.1177/102986499700100205

Juslin, P. N., & Västfjäll, D. (2008). Emotional responses to music: The need to consider underlying mechanisms. Behavioral and brain sciences, 31(5), 559-575. https://doi-org.ezproxy.uio.no/10.1017/S0140525X08005293

Kallinen, K. (2002). Reading news from a pocket computer in a distracting environment: effects of the tempo of background music. Computers in Human Behavior, 18(5), 537- 551. https://doi.org/10.1016/S0747-5632(02)00005-5

Karageorghis, C. I., Jones, L., & Low, D. C. (2006). Relationship between exercise heart rate and music tempo preference. Research quarterly for exercise and sport, 77(2), 240-250. https://pubmed.ncbi.nlm.nih.gov/16898279/

Kastner, M. P. & Crowder, R. G. (1990). Perception of the major/minor distinction: IV. Emotional connotations in young children. Music Perception, 8(2), 189-201. https://doi.org/10.2307/40285496

Kellaris, J. J. & Rice, R. C. (1993). The influence of tempo, loudness, and gender of listener on responses to music. Psychology & Marketing, 10(1), 15-29. https://doi-org.ezproxy.uio.no/10.1002/mar.4220100103

Khalfa, S., Roy, M., Rainville, P., Dalla Bella, S., & Peretz, I. (2008). Role of tempo entrainment in psychophysiological differentiation of happy and sad music?. International Journal of Psychophysiology, 68(1), 17-26. doi: 10.1097/00001756-200512190-00002

Kirschner, S. & Tomasello, M. (2010). Joint music making promotes prosocial behavior in 4- year-old children. Evolution and Human Behavior, 31(5), 354-364. https://doi.org/10.1016/j.evolhumbehav.2010.04.004

Kistler, M. E. & Lee, M. J. (2009). Does exposure to sexual hip-hop music videos influence the sexual attitudes of college students?. Mass Communication and Society, 13(1), 67- 86. https://doi-org.ezproxy.uio.no/10.1080/15205430902865336

Klohnen, E. C. & Luo, S. (2003). Interpersonal attraction and personality: What is attractive-- self similarity, ideal similarity, complementarity or attachment security?. Journal of personality and social psychology, 85(4), 709. https://doi-org.ezproxy.uio.no/10.1037/0022-3514.85.4.709

Koelsch, S., Fritz, T., v. Cramon, D. Y., Müller, K., & Friederici, A. D. (2006). Investigating emotion with music: an fMRI study. Human brain mapping, 27(3), 239-250. https://doi-org.ezproxy.uio.no/10.1002/hbm.20180

Konvalinka, I., Vuust, P., Roepstorff, A., & Frith, C. D. (2010). Follow you, follow me: continuous mutual prediction and adaptation in joint tapping. Quarterly journal of experimental psychology, 63(11), 2220-2230. https://doi-org.ezproxy.uio.no/10.1080/17470218.2010.497843

Kreutz, G. (2014). Does singing facilitate social bonding. Music Med, 6(2), 51-60. http://mmd.iammonline.com/index.php/musmed/article/view/MMD-6-2-9

Kreutz, G., Ott, U., Teichmann, D., Osawa, P., & Vaitl, D. (2008). Using music to induce emotions: Influences of musical preference and absorption. Psychology of music, 36(1), 101-126. https://doi-org.ezproxy.uio.no/10.1177/0305735607082623

Kteily, N., Bruneau, E., Waytz, A., & Cotterill, S. (2015). The ascent of man: Theoretical and empirical evidence for blatant dehumanization. Journal of personality and social psychology, 109(5), 901. https://doi-org.ezproxy.uio.no/10.1037/pspp0000048

Kuznetsova, A., Brockhoff, P.B. & Christensen, R.H.B. (2017). “lmerTest Package: Tests in Linear Mixed Effects Models.” Journal of Statistical Software, 82(13), 1–26. doi: 10.18637/jss.v082.i13.

Ladinig, O. & Schellenberg, E. G. (2012). Liking unfamiliar music: Effects of felt emotion and individual differences. Psychology of Aesthetics, Creativity, and the Arts, 6(2), 146. https://doi-org.ezproxy.uio.no/10.1037/a0024671

LeBlanc, A., Colman, J., McCrary, J., Sherrill, C., & Malin, S. (1988). Tempo preferences of different age music listeners. Journal of research in music education, 36(3), 156-168. https://doi-org.ezproxy.uio.no/10.2307/3344637

Lennings, H. I. B. & Warburton, W. A. (2011). The effect of auditory versus visual violent media exposure on aggressive behaviour: The role of song lyrics, video clips and musical tone. Journal of Experimental Social Psychology, 47(4), 794-799. https://doi.org/10.1016/j.jesp.2011.02.006

Lev-Ari, S. & Keysar, B. (2010). Why don't we believe non-native speakers? The influence of accent on credibility. Journal of experimental social psychology, 46(6), 1093-1096. https://doi.org/10.1016/j.jesp.2010.05.025

Levin, S., Van Laar, C., & Sidanius, J. (2003). The effects of ingroup and outgroup friendships on ethnic attitudes in college: A longitudinal study. Group Processes & Intergroup Relations, 6(1), 76-92. https://doi-org.ezproxy.uio.no/10.1177/1368430203006001013

Lonsdale, A. J. & North, A. C. (2009). Musical taste and ingroup favouritism. Group Processes & Intergroup Relations, 12(3), 319-327. https://doi-org.ezproxy.uio.no/10.1177/1368430209102842

Maher, P. J., Van Tilburg, W. A. P., & Van Den Tol, A. J. M. (2013). Meaning in music: Deviations from expectations in music prompt outgroup derogation. European Journal of Social Psychology, 43(6), 449-454. https://doi-org.ezproxy.uio.no/10.1002/ejsp.1969

Margulis, E. H. (2005). A model of melodic expectation. Music Perception: An Interdisciplinary Journal, 22(4), 663-714. https://doi.org/10.1525/mp.2005.22.4.663

Masataka, N. (2006). Preference for consonance over dissonance by hearing newborns of deaf parents and of hearing parents. Developmental Science, 9, 46–50. doi:10.1111/j.1467-7687.2005.00462.x

Mazziotta, A., Mummendey, A., & Wright, S. C. (2011). Vicarious intergroup contact effects: Applying social-cognitive theory to intergroup contact research. Group Processes & Intergroup Relations, 14(2), 255-274. https://doi-org.ezproxy.uio.no/10.1177/1368430210390533

McAuley, J. D., Jones, M. R., Holub, S., Johnston, H. M., & Miller, N. S. (2006). The time of our lives: life span development of timing and event tracking. Journal of Experimental Psychology: General, 135(3), 348. https://doi-org.ezproxy.uio.no/10.1037/0096-3445.135.3.348

McDermott, J. H., Lehr, A. J., & Oxenham, A. J. (2010). Individual differences reveal the basis of consonance. Current Biology, 20(11), 1035-1041. https://doi.org/10.1016/j.cub.2010.04.019

McDermott, J. H., Schultz, A. F., Undurraga, E. A., & Godoy, R. A. (2016). Indifference to dissonance in native Amazonians reveals cultural variation in music perception. Nature, 535(7613), 547-550. https://doi-org.ezproxy.uio.no/10.1038/nature18635

Mehr, S. A., Singh, M., Knox, D., Ketter, D. M., Pickens-Jones, D., Atwood, S., ... & Howard, R. M. (2019). Universality and diversity in human song. Science, 366(6468). https://doi.org/10.1016/j.cub.2017.12.042

Merker, B. (2000) Synchronous chorusing and human origins. In The Origins of Music, N. L. Wallin, B. Merker, and S. Brown, eds. (pp. 315-327). MIT Press.

Moelants, D. (2003). Dance music, movement and tempo preferences. In Proceedings of the 5th Triennial ESCOM Conference (pp. 649-652). Hanover University of Music and Drama. http://hdl.handle.net/1854/LU-213437

Laham, S. M., Koval, P., & Alter, A. L. (2012). The name-pronunciation effect: Why people like Mr. Smith more than Mr. Colquhoun. Journal of Experimental Social Psychology, 48(3), 752-756. https://doi.org/10.1016/j.jesp.2011.12.002

Lartillot, O. & Toiviainen, P. (2007). A Matlab toolbox for musical feature extraction from audio. In International conference on digital audio effects (pp. 237-244). www.jyu.fi/music/coe/materials/mirtoolbox

Lee, D., Park, J. Y., Kim, J., Kim, J., & Moon, J. (2011). Understanding music sharing behaviour on social network services. Online Information Review. 35(5), 716-733. https://doi.org/10.1108/14684521111176462

Lennings, H. I. B., & Warburton, W. A. (2011). The effect of auditory versus visual violent media exposure on aggressive behaviour: The role of song lyrics, video clips and musical tone. Journal of Experimental Social Psychology, 47(4), 794-799. https://doi.org/10.1016/j.jesp.2011.02.006

Lick, D. J. & Johnson, K. L. (2013). Fluency of visual processing explains prejudiced evaluations following categorization of concealable identities. Journal of Experimental Social Psychology, 49(3), 419-425. https://doi-org.ezproxy.uio.no/10.1521/soco.2013.31.6.656

Lick, D. J. & Johnson, K. L. (2015). The interpersonal consequences of processing ease: Fluency as a metacognitive foundation for prejudice. Current Directions in Psychological Science, 24(2), 143-148. https://doi-org.ezproxy.uio.no/10.1177/0963721414558116

Omar, R., Henley, S. M., Bartlett, J. W., Hailstone, J. C., Gordon, E., Sauter, D. A., ... & Warren, J. D. (2011). The structural neuroanatomy of music emotion recognition: evidence from frontotemporal lobar degeneration. Neuroimage, 56(3), 1814-1821. https://doi.org/10.1016/j.neuroimage.2011.03.002

Pallesen, K. J., Brattico, E., Bailey, C., Korvenoja, A., Koivisto, J., Gjedde, A., & Carlson, S. (2005). Emotion processing of major, minor, and dissonant chords: a functional magnetic resonance imaging study. Annals of the New York Academy of Sciences, 1060(1), 450-453. doi: 10.1196/annals.1360.047

Panksepp, J. & Bernatzky, G. (2002). Emotional sounds and the brain: the neuro-affective foundations of musical appreciation. Behavioural processes, 60(2), 133-155. https://doi.org/10.1016/S0376-6357(02)00080-3

Park, M., Gutyrchik, E., Bao, Y., Zaytseva, Y., Carl, P., Welker, L., ... & Meindl, T. (2014). Differences between musicians and non-musicians in neuro-affective processing of sadness and fear expressed in music. Neuroscience letters, 566, 120-124. https://doi.org/10.1016/j.neulet.2014.02.041

Pearson, A. R. (2011). Intergroup fluency: A metacognitive approach to intergroup perception. Yale University.

Peretz, I., Gagnon, L., & Bouchard, B. (1998). Music and emotion: perceptual determinants, immediacy, and isolation after brain damage. Cognition, 68(2), 111-141. https://doi.org/10.1016/S0010-0277(98)00043-2

Pinker, S. (2010). The cognitive niche: Coevolution of intelligence, sociality, and language. Proceedings of the National Academy of Sciences, 107(Supplement 2), 8993- 8999. https://doi-org.ezproxy.uio.no/10.1073/pnas.0914630107

Plant, E. A., & Devine, P. G. (2003). The antecedents and implications of interracial anxiety. Personality and Social Psychology Bulletin, 29(6), 790-801. https://doi-org.ezproxy.uio.no/10.1177/0146167203029006011

Plantinga, J. & Trehub, S. E. (2014). Revisiting the innate preference for consonance. Journal of Experimental Psychology: Human Perception and Performance, 40(1), 40. https://doi-org.ezproxy.uio.no/10.1037/a0033471

Plomp, R. & Levelt, W. (1965). Tonal consonance and critical bandwidth. Journal of the Acoustical Society of America, 38, 548 –560. doi:10.1121/ 1.1909741

Quinto, L. R., Thompson, W. F., & Keating, F. L. (2013). Emotional communication in speech and music: The role of melodic and rhythmic contrasts. Frontiers in psychology, 4, 184. https://doi.org/10.3389/fpsyg.2013.00184

Rabinowitch, T. C., Cross, I., & Burnard, P. (2013). Long-term musical group interaction has a positive influence on empathy in children. Psychology of music, 41(4), 484-498. https://doi-org.ezproxy.uio.no/10.1177/0305735612440609

Reber, R., Wurtz, P., & Zimmermann, T. D. (2004). Exploring “fringe” consciousness: The subjective experience of perceptual fluency and its objective bases. Consciousness and cognition, 13(1), 47-60. https://doi.org/10.1016/S1053-8100(03)00049-7

Rentfrow, P. J. & Gosling, S. D. (2006). Message in a ballad: The role of music preferences in interpersonal perception. Psychological science, 17(3), 236-242. https://doi-org.ezproxy.uio.no/10.1111/j.1467-9280.2006.01691.x

Rigg, M. G. (1940). Speed as a determiner of musical mood. Journal of Experimental Psychology, 27(5), 566. https://doi-org.ezproxy.uio.no/10.1037/h0058652

Roccas, S. & Schwartz, S. H. (1993). Effects of intergroup similarity on intergroup relations. European Journal of Social Psychology, 23(6), 581-595. https://doi-org.ezproxy.uio.no/10.1002/ejsp.2420230604

Roederer, J. G. (1984). The search for a survival value of music. Music Perception: An Interdisciplinary Journal, 1(3), 350-356. https://doi.org/10.2307/40285265

Rodríguez-Bailón, R., Ruiz, J., & Moya, M. (2009). The impact of music on automatically activated attitudes: Flamenco and gypsy people. Group Processes & Intergroup Relations, 12(3), 381-396. https://doi-org.ezproxy.uio.no/10.1177/1368430209102849

Ross, D., Choi, J., & Purves, D. (2007). Musical intervals in speech. Proceedings of the National Academy of Sciences, 104(23), 9852-9857. https://doi-org.ezproxy.uio.no/10.1073/pnas.0703140104

Rossing, T. D., Moore, R., & Wheeler, P. (2002). The science of sound (3rd ed.). Pearson Education, Inc.

Rosenfeld, A. H. (1985). Music, the beautiful disturber. Psychology Today, 19(12), 48.

Russo, F. A. & Thompson, W. F. (2005). The subjective size of melodic intervals over a two- octave range. Psychonomic bulletin & review, 12(6), 1068-1075. https://link-springer-com.ezproxy.uio.no/article/10.3758/BF03206445

Schiappa, E., Gregg, P. B., & Hewes, D. E. (2005). The parasocial contact hypothesis. Communication monographs, 72(1), 92-115. https://doi.org/10.1080/0363775052000342544

Schlueter, E., & Scheepers, P. (2010). The relationship between outgroup size and anti- outgroup attitudes: A theoretical synthesis and empirical test of group threat-and intergroup contact theory. Social Science Research, 39(2), 285-295. https://doi.org/10.1016/j.ssresearch.2009.07.006

Schlueter, E., Ullrich, J., Glenz, A., & Schmidt, P. (2018). From segregation to intergroup contact and back: Using experiments and simulation to understand the bidirectional link. European Journal of Social Psychology, 48(1), 17-32. https://doi-org.ezproxy.uio.no/10.1002/ejsp.2284

Schmid, K., Ramiah, A. A., & Hewstone, M. (2014). Neighborhood ethnic diversity and trust: The role of intergroup contact and perceived threat. Psychological science, 25(3), 665- 674. https://doi-org.ezproxy.uio.no/10.1177/0956797613508956

Schwartz, D. A., Howe, C. Q., & Purves, D. (2003). The statistical structure of human speech sounds predicts musical universals. Journal of Neuroscience, 23(18), 7160-7168. https://doi-org.ezproxy.uio.no/10.1523/JNEUROSCI.23-18-07160.2003

Simonton, D. K. (1980a). Thematic fame and melodic originality in classical music: A multivariate computer‐content analysis 1. Journal of Personality, 48(2), 206-219. https://doi-org.ezproxy.uio.no/10.1111/j.1467-6494.1980.tb00828.x

Simonton, D. K. (1980b). Thematic fame, melodic originality, and musical zeitgeist: A biographical and transhistorical content analysis. Journal of Personality and Social Psychology, 38(6), 972–983. https://doi-org.ezproxy.uio.no/10.1037/0022-3514.38.6.972

Singh, R. & Ho, S. Y. (2000). Attitudes and attraction: A new test of the attraction, repulsion and similarity‐dissimilarity asymmetry hypotheses. British Journal of Social Psychology, 39(2), 197-211. https://doi-org.ezproxy.uio.no/10.1348/014466600164426

Stephan, W. G., Diaz-Loving, R., & Duran, A. (2000). Integrated threat theory and intercultural attitudes: Mexico and the United States. Journal of cross-cultural Psychology, 31(2), 240-249. https://doi-org.ezproxy.uio.no/10.1177/0022022100031002006

Stephan, W. G., & Stephan, C. W. (1985). Intergroup anxiety. Journal of social issues, 41(3), 157-175. https://doi-org.ezproxy.uio.no/10.1111/j.1540-4560.1985.tb01134.x

Stephan, W. G., Stephan, C. W., & Gudykunst, W. B. (1999). Anxiety in intergroup relations: A comparison of anxiety/uncertainty management theory and integrated threat theory. International Journal of Intercultural Relations, 23(4), 613-628. https://doi.org/10.1016/S0147-1767(99)00012-7

Stroebe, W., Insko, C. A., Thompson, V. D., & Layton, B. D. (1971). Effects of physical attractiveness, attitude similarity, and sex on various aspects of interpersonal attraction. Journal of Personality and Social Psychology, 18(1), 79. https://doi-org.ezproxy.uio.no/10.1037/h0030710

Struch, N., & Schwartz, S. H. (1989). Intergroup aggression: Its predictors and distinctness from in-group bias. Journal of personality and social psychology, 56(3), 364. https://doi-org.ezproxy.uio.no/10.1037/0022-3514.56.3.364

Taguchi, N. (2007). Development of speed and accuracy in pragmatic comprehension in English as a foreign language. Tesol Quarterly, 41(2), 313-338. https://doi-org.ezproxy.uio.no/10.1002/j.1545-7249.2007.tb00061.x

Tarr, B., Launay, J., Cohen, E., & Dunbar, R. (2015). Synchrony and exertion during dance independently raise pain threshold and encourage social bonding. Biology letters, 11(10), 20150767. https://doi-org.ezproxy.uio.no/10.1098/rsbl.2015.0767

Tekman, H. G., & Hortaçsu, N. (2002). Music and social identity: Stylistic identification as a response to musical style. International Journal of Psychology, 37(5), 277-285. https://doi-org.ezproxy.uio.no/10.1080/00207590244000043

Toiviainen, P. & Eerola, T. (2016). MIDI Toolbox 1.1. Github. https://github.com/miditoolbox/1.1

Trainor, L. J. & Heinmiller, B. M. (1998). The development of evaluative responses to music: Infants prefer to listen to consonance over dissonance. Infant Behavior & Development, 21, 77– 88. doi:10.1016/S0163- 6383(98)90055-8

Trochidis, K. & Bigand, E. (2013). Investigation of the effect of mode and tempo on emotional responses to music using EEG power asymmetry. Journal of Psychophysiology. https://doi-org.ezproxy.uio.no/10.1027/0269-8803/a000099.

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Basil Blackwell.

Van de Geer, J., Levelt W., Plomp, R. (1962): The connotation of musical consonance. Acta Psychol 20: 308– 319. https://doi.org/10.1016/0001-6918(62)90028-8

Van der Zwaag, M. D., Westerink, J. H., & Van den Broek, E. L. (2011). Emotional and psychophysiological responses to tempo, mode, and percussiveness. Musicae Scientiae, 15(2), 250-269. https://doi-org.ezproxy.uio.no/10.1177/1029864911403364

Vassilakis, P. N. (2005). Auditory roughness as a means of musical expression. Selected Reports in Ethnomusicology, 12(119-144), 122.

Virtala, P. & Tervaniemi, M. (2017). Neurocognition of major-minor and consonance- dissonance. Music Perception: An Interdisciplinary Journal, 34(4), 387-404. https://doi.org/10.1525/mp.2017.34.4.387

Vuoskoski, J. K., Clarke, E. F., & DeNora, T. (2017). Music listening evokes implicit affiliation. Psychology of Music, 45(4), 584-599. https://doi-org.ezproxy.uio.no/10.1177/0305735616680289

Vuoskoski, J. K. & Eerola, T. (2012). Can sad music really make you sad? Indirect measures of affective states induced by music and autobiographical memories. Psychology of Aesthetics, Creativity, and the Arts, 6(3), 204. https://doi-org.ezproxy.uio.no/10.1037/a0026937

Wachsmann, K. P. (1971). Universal perspectives in music. Ethnomusicology, 15(3), 381- 384. https://www-jstor-org.ezproxy.uio.no/stable/850638

Wagner, U., Christ, O., Pettigrew, T. F., Stellmacher, J., & Wolf, C. (2006). Prejudice and minority proportion: Contact instead of threat effects. Social psychology quarterly, 69(4), 380-390. https://doi-org.ezproxy.uio.no/10.1177/019027250606900406

Webster, G. D. & Weir, C. G. (2005). Emotional responses to music: Interactive effects of mode, texture, and tempo. Motivation and Emotion, 29(1), 19-39. https://link-springer-com.ezproxy.uio.no/article/10.1007/s11031-005-4414-0

Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York. ISBN 978-3-319-24277-4, https://ggplot2.tidyverse.org.

Winkielman, P., Halberstadt, J., Fazendeiro, T., & Catty, S. (2006). Prototypes are attractive because they are easy on the mind. Psychological science, 17(9), 799-806. https://doi-org.ezproxy.uio.no/10.1111/j.1467-9280.2006.01785.x

Williams, J (1975). Main Title (Theme From 'Jaws'). [Recorded by MCA (Boston Pops Orchestra)]. Jaws: Original Motion Picture Soundtrack [LP]. MCA Records.

Wright, S. C., Aron, A., McLaughlin-Volpe, T., & Ropp, S. A. (1997). The extended contact effect: Knowledge of cross-group friendships and prejudice. Journal of Personality and Social psychology, 73(1), 73. doi:10.1037/0022-3514.73.1.73

Zhou, S., Page-Gould, E., Aron, A., Moyer, A., & Hewstone, M. (2019). The extended contact hypothesis: A meta-analysis on 20 years of research. Personality and Social Psychology Review, 23(2), 132-160. https://doi-org.ezproxy.uio.no/10.1177/1088868318762647

Ziegler, C. N. & Golbeck, J. (2007). Investigating interactions of trust and interest similarity. Decision support systems, 43(2), 460-475. https://doi.org/10.1016/j.dss.2006.11.003

Appendix A

Study 1 instructions and rating scales

Part 1

Thank you for choosing to participate in this study.

This study has 2 parts in which you will be presented with 60 music excerpts. In the first part you will be presented 30 music excerpts. After listening to each music excerpt your task is to rate each of these melodies, using multiple choice and the slider bars on eight domains.

Please remember to use headphones in order to minimize input from other sounds nearby.

The domains are: Warmth - how warm/affectionate/kind does the music sound? Advancement - how advanced does the music sound? Threat - how threatening does the music sound? Energy - how energetic is the music? Liking - how much do you like the music? Happiness - how happy does the music sound? Familiarity - how familiar does the music seem? Geographical region - where in the world do you think the melody is from?

Please click the play button to listen to the song. When the song is finished, answer the following questions.

How warm/affectionate/kind does the music sound?

Cold Warm

How advanced does the music sound?

Primitive Advanced

How threatening does the music sound?

Non -threatening Threatening

How energetic does the music sound?

Non-energetic Energetic

How much do you like the music?

Dislike Like

How happy does the music sound?

Not happy at all Very happy

How familiar does the music seem?

Unfamiliar Familiar

Where in the world do you think the music is from? • Africa • Oceania • East Asia • South Asia • Middle East

• Eastern Europe • Western Europe • North America • South America

Part 2

Thank you for completing the first part of this study. The music excerpts you listened to actually came from different, relatively unknown cultures. In the next part of the study, we would like you to listen to the excerpts once more, this time rating the culture they come from, based on the music you hear. Try to base your ratings on what the music itself conveys, and not where you think the music comes from.

The domains are: Warmth - how warm/affectionate/kind does the culture seem? Competence - how competent does the culture seem? Threat - how threatening does the culture seem? Liking - how much do you think you would like the culture? Happiness - how happy does the culture seem? Evolved - how evolved do you consider the culture to be?

Please click the play button to listen to the song. When the song is finished, answer the following questions.

How warm/affectionate/kind does the culture seem?

Cold Warm

How competent does the culture seem?

Primitive Advanced

How threatening does the culture seem?

Non-threatening Threatening

How much do you think you would like the culture?

Dislike Like

How happy does the culture seem?

Not happy at all Very happy

Using the slider below, indicate how evolved you consider the culture to be.

Study 2 instructions:

Thank you for choosing to participate in this study.

In this study you will be presented 6 music excerpts from various relatively unknown cultures. After listening to each music excerpt your task is to rate each of these melodies based on your impression of the culture the music stems from, using the slider bars on six domains.

Try to base your ratings on what the music itself conveys, and not on any existing cultures that the music may remind you of.

Please remember to use headphones in order to minimize input from other sounds nearby.

(Rating scales are equal to Study 1, Part 2)

Appendix B

Linear mixed model equations from Study 1:

Figure A1

Models 1-7: Equation of Musical Ratings on Musical Parameters:

!!" = a + ap + as + (bp1 + b1)x1i + (bp2 + b2)x2i +…+(bp5 + b5)x5i + eij

Note. !!" is the dependent variable (musical rating dimension). a is the intercept. ap is the random intercept for participant. as is the random intercept for song. bp1 is the slope of the random effect of predictor 1(musical parameter) by participant. x1i is the value of the effect of predictor 1. b1 is the regular slope for predictor 1. eij is the remaining random variance. No random slopes for predictors by song, as predictors (musical parameters) are between song variables.

Figure A2

Model 8-11: Equation of Cultural Ratings on Musical Parameters:

!!" = a + ap + as + (bp1 + b1)x1i + (bp2 + b2)x2i +…+(bp5 + b5)x5i + eij

Model Interaction of Musical Happiness on General Pitch Variation and Interval Size:

!!" = a + ap + as + (bp1 + b1)x1i + (bp2 + b2)x2i +…+(bp5 + b5)x5i + b6x2i x3i + eij

Model Interaction of Cultural Warmth on General Pitch Variation and Interval Size:

!!" = a + ap + as + (bp1 + b1)x1i + (bp2 + b2)x2i +…+(bp5 + b5)x5i + b6x2i x3i + eij

Note. !!" is the dependent variable (musical rating dimension). a is the intercept. ap is the random intercept for participant. as is the random intercept for song. bp1 is the slope of the random effect of predictor 1(musical parameter) by participant. x1i is the value of the effect of predictor 1. b1 is the regular slope for predictor 1. b6 is the slope of the interaction. x2i is the value of the random effect of general pitch variation. x3i is the value of the random effect of interval size. No random slopes for predictors by song, as predictors (musical parameters) are between song variables. eij is the remaining random variance.

Figure A3

Models 12-15: Equation of Cultural Ratings on Musical Ratings:

!!" = a + ap + as + (bp1 + b1)x1i + (bp2 + b2)x2i +…+(bp7 + b7)x7i + (bs1 + b1)x1j + (bs2 + b2)x2j +…+(bs7 + b7)x7j + eij

Note. !!" is the dependent variable (cultural evaluation dimension). a is the intercept. ap is the random intercept for participant. as is the random intercept for song. bp1 is the slope of the random effect of predictor 1(musical perception dimensions) by participant. x1i is the value of the effect of predictor 1 by participant. bs1 is the slope of the random effect of predictor 1 by song. x1j is the value of the effect of predictor 1 by song. b1 is the regular slope for predictor 1. eij is the remaining random variance.

Linear mixed model equations from Study 2:

Figure A4

Models 16-19: Equation of Cultural Ratings on Musical Parameters:

!!" = a + ap + as + (bp1 + b1)x1i + (bp2 + b2)x2i + (bs1 + b1)x1j + (bs2 + b2)x2j + b3x1ijx2ij+ eij

Note. !!" is the dependent variable (cultural evaluation dimension). a is the intercept. ap is the random intercept for participant. as is the random intercept for song. bp1 is the slope of the random effect of predictor 1(musical parameter) by participant. x1i is the value of the effect of predictor 1 by participant. bs1 is the slope of the random effect of predictor 1 by song. x1j is the value of the effect of predictor 1 by song. b1 is the regular slope for predictor 1. b3 is the slope of the interaction effect between parameters. eij is the remaining random variance.

Figure A5

Models 20-23: Equation of Individual Differences on Cultural Ratings

!!" = a + ap + as + (bs1 + b1)x1j + (bs2 + b2)x2j + (bs3 + b3)x3j + (bs4 + b4)x4j + (bp1 + b4)x4i + b6x3jx4ij + b7x2jx3j + eij

Note. !!" is the dependent variable (cultural evaluation dimension). a is the intercept. ap is the random intercept for participant. as is the random intercept for song. bs1 is the slope of the random effect of predictor 1(gender) by song. x1j is the value of the effect of predictor 1 by song. b1 is the regular slope for predictor 1. b4 is the regular slope for dissonance. b6 is the slope of the interaction effect between musical training and dissonance. b7 is the slope of the interaction effect between age and musical training. eij is the remaining random variance. No random slopes for individual differences by participant, as individual differences are between participant variables.

Figure A6

Models 24-25: Equation of Individual Differences on Cultural Ratings: Liking and Happiness

!!" = a + ap + as + (bs1 + b1)x1j + (bs2 + b2)x2j + (bs3 + b3)x3j + eij

Figure A7

Models 26-29: Equation of Cultural Ratings on Amount of Dissonance

!!" = a + ap + as + (bp1 + b1)x1i + (bp2 + b2)x2i + eij

Note. !!" is the dependent variable (cultural evaluation dimension). a is the intercept. ap is the random intercept for participant. as is the random intercept for song. bp1 is the slope of the random effect of predictor 1(percentage of dissonant intervals) by participant. x1i is the value of the effect of predictor 1 by participant. b1 is the regular slope for predictor 1. eij is the remaining random variance. No random slopes for predictors by song, as predictors (amount of dissonance) are between song variables.

Appendix C Tables with regression coefficient (unstandardized dependent variables)

Table C1 Cultural Warmth on Musical Ratings Fixed effects Estimate Std. Error t value Pr(>|t|) (Intercept) 56.3110 1.0773 52.271 < 2e-16 *** W1 2.4095 1.0717 2.248 0.029055 * T1 -4.0683 0.6632 -6.134 4.65e-09 *** L1 1.3562 0.8199 1.654 0.102393 E1 3.4903 0.8865 3.937 0.000192 *** H1 3.4122 0.9979 3.419 0.000973 *** A1 1.7458 0.7495 2.329 0.023941 * F1 -0.4106 0.6809 -0.603 0.547740 Note. W1=Musical warmth, T1=Musical threat, L1=Musical liking, E1=Musical energy, H1=Musical happiness, A1=Musical advancement, F1=Musical familiarity.

Table C2 Cultural Threat on Musical Ratings Fixed effects Estimate Std. Error t value Pr(>|t|) (Intercept) 28.2194 1.4772 19.104 < 2e-16 *** W1 -3.1034 1.1321 -2.741 0.00858 ** T1 7.1457 0.8631 8.279 1.67e-10 *** L1 -0.9721 0.8841 -1.099 0.27728 E1 -1.5235 0.9959 -1.530 0.13181 H1 0.4705 0.9517 0.494 0.62178 A1 -0.6391 0.7568 -0.844 0.40175 F1 0.6878 0.7761 0.886 0.37869 Note. W1=Musical warmth, T1=Musical threat, L1=Musical liking, E1=Musical energy, H1=Musical happiness, A1=Musical advancement, F1=Musical familiarity.

Table C3 Cultural Competence on Musical Ratings Fixed effects Estimate Std. Error t value Pr(>|t|) (Intercept) 62.3523 1.6449 37.906 < 2e-16 *** W1 -0.6045 0.7980 -0.757 0.450102 T1 -2.8182 0.8110 -3.475 0.001159 ** L1 3.2049 0.7866 4.074 0.000192 *** E1 3.8659 0.9783 3.952 0.000312 *** H1 0.9959 0.8801 1.132 0.259765 A1 2.7921 0.7358 3.795 0.000495 *** F1 1.5932 0.8339 1.910 0.061825 . Note. W1=Musical warmth, T1=Musical threat, L1=Musical liking, E1=Musical energy, H1=Musical happiness, A1=Musical advancement, F1=Musical familiarity.

Table C4 Cultural Evolvedness on Musical Ratings Fixed effects Estimate Std. Error t value Pr(>|t|) (Intercept) 79.7269 2.3806 33.491 < 2e-16 *** W1 -0.3582 0.6469 -0.554 0.58134 T1 -2.2080 0.6396 -3.452 0.00129 ** L1 2.3458 0.6291 3.729 0.00044 *** E1 2.1879 0.7832 2.794 0.00773 ** H1 1.6472 0.9253 1.780 0.08033 . A1 1.5698 0.6140 2.557 0.01453 * F1 0.2428 0.5985 0.406 0.68657 Note. W1=Musical warmth, T1=Musical threat, L1=Musical liking, E1=Musical energy, H1=Musical happiness, A1=Musical advancement, F1=Musical familiarity.

Table C5 Cultural Warmth on Musical parameters Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 56.2256 1.2180 46.163 < 2e-16 *** Org 1.3799 0.6796 2.030 0.042485 * GPV -1.5712 0.5529 -2.841 0.004551 ** Int 3.0116 0.8344 3.609 0.000317 *** MusVel 8.6586 0.7922 10.930 < 2e-16 *** Mode 3.1647 0.5082 6.227 6.15e-10 *** GPV:Int -1.9899 0.5123 -3.884 0.000107 *** Note. Org=Originality, GPV = General Pitch variation, Int = Interval size, MusVel = Musical Velocity, Mode= Modality

Table C6 Cultural Threat on Musical Parameters Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 27.6500 1.8816 14.695 < 2e-16 *** Org 0.3135 0.7095 0.442 0.658665 GPV 0.7253 0.5772 1.256 0.209136 Int -2.4563 0.8710 -2.820 0.004863 ** MusVel -2.9489 0.8270 -3.566 0.000374 *** Mode -3.6631 0.5305 -6.905 7.4e-12 *** Note. Org=Originality, GPV = General Pitch variation, Int = Interval size, MusVel = Musical Velocity, Mode= Modality

Table C7 Cultural Competentence on Musical Parameters Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 63.1839 1.8567 34.029 < 2e-16 *** Org 2.3257 0.6334 3.672 0.000249 *** GPV 0.7646 0.5153 1.484 0.138100 Int 1.6116 0.7776 2.073 0.038386 * MusVel 5.4708 0.7383 7.410 2.1e-13 *** Mode 0.6945 0.4736 1.466 0.142803 Note. Org=Originality, GPV = General Pitch variation, Int = Interval size, MusVel = Musical Velocity, Mode= Modality

Table C8 Cultural Evolvedness on Musical Parameters Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 80.6487 2.2844 35.305 < 2e-16 *** Org 1.4066 0.4951 2.841 0.00456 ** GPV 0.5314 0.4028 1.319 0.18732 Int 0.7982 0.6078 1.313 0.18932 MusVel 3.5601 0.5771 6.169 8.83e-10 *** Mode 0.7843 0.3702 2.118 0.03430 * Note. Org=Originality, GPV = General Pitch variation, Int = Interval size, MusVel = Musical Velocity, Mode= Modality

Table C9 Musical Warmth on Musical Parameters Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 53.0731 1.1885 44.655 < 2e-16 *** Org 1.0358 0.6744 1.536 0.125 GPV -0.6519 0.5487 -1.188 0.235 Int 1.3282 0.8280 1.604 0.109 MusVel 8.4264 0.7861 10.719 < 2e-16 *** Mode 3.6363 0.5043 7.210 8.81e-13 *** Note. Org=Originality, GPV = General Pitch variation, Int = Interval size, MusVel = Musical Velocity, Mode= Modality

Table C10 Musical Threat on Musical Parameters Fixed effects Estimate Std. Error t value Pr(>|t|) (Intercept) 23.2942 1.7692 13.166 < 2e-16 *** Org -0.2075 0.6444 -0.322 0.7475 GPV 0.2852 0.5243 0.544 0.5865 Int -1.4325 0.7911 -1.811 0.0704 . MusVel -4.3641 0.7511 -5.810 7.62e-09 *** Mode -3.4350 0.4819 -7.128 1.57e-12 *** Note. Org=Originality, GPV = General Pitch variation, Int = Interval size, MusVel = Musical Velocity, Mode= Modality

Table C11 Musical Energy on Musical Parameters Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 47.9391 1.0973 43.689 < 2e-16 *** Org 1.6333 0.6604 2.473 0.013507 * GPV -0.8453 0.5373 -1.573 0.115922 Int 3.1398 0.8108 3.872 0.000112 *** MusVel 14.1090 0.7698 18.327 < 2e-16 *** Mode 1.2573 0.4939 2.546 0.011002 * Note. Org=Originality, GPV = General Pitch variation, Int = Interval size, MusVel = Musical Velocity, Mode= Modality

Table C12 Musical Liking on Musical Parameters Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 54.90508 1.64935 33.289 < 2e-16 *** Org 3.06470 0.63663 4.814 1.63e-06 *** GPV 0.55108 0.51797 1.064 0.2875 Int 1.78432 0.78159 2.283 0.0226 * MusVel 6.30647 0.74208 8.498 < 2e-16 *** Mode 0.05282 0.47606 0.111 0.9117 Note. Org=Originality, GPV = General Pitch variation, Int = Interval size, MusVel = Musical Velocity, Mode= Modality

Table C13 Musical Happiness on Musical Parameters Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 47.2231 1.2402 38.076 < 2e-16 *** Org 0.8104 0.7116 1.139 0.25493 GPV -1.4366 0.5790 -2.481 0.01320 * Int 2.6497 0.8736 3.033 0.00246 ** MusVel 12.3444 0.8295 14.882 < 2e-16 *** Mode 2.8712 0.5321 5.396 7.92e-08 *** GPV:Int -0.4041 0.5390 -0.750 0.45361 Note. Org=Originality, GPV = General Pitch variation, Int = Interval size, MusVel = Musical Velocity, Mode= Modality

Table C14 Musical Advancement on Musical Parameters Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 50.4891 1.6833 29.995 < 2e-16 *** Org 2.1994 0.6281 3.502 0.000476 *** GPV 1.1889 0.5110 2.327 0.020123 * Int 1.6562 0.7711 2.148 0.031887 * MusVel 5.9579 0.7321 8.138 8.37e-16 *** Mode 0.9546 0.4697 2.032 0.042296 * Note. Org=Originality, GPV = General Pitch variation, Int = Interval size, MusVel = Musical Velocity, Mode= Modality

Table C15 Musical Familiarity on Musical Parameters Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 40.1878 2.5631 15.679 < 2e-16 *** Org 2.0676 0.6770 3.054 0.00230 ** GPV 0.8406 0.5508 1.526 0.12716 Int 0.3842 0.8311 0.462 0.64397 MusVel 3.4295 0.7891 4.346 1.48e-05 *** Mode -1.2351 0.5062 -2.440 0.01481 * Note. Org=Originality, GPV = General Pitch variation, Int = Interval size, MusVel = Musical Velocity, Mode= Modality

Table C16 Cultural Warmth on Dissonance and Tempo

Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 61.5652 2.1103 29.173 1.42e-08 *** Dissonant -13.1499 1.9884 -6.613 0.000408 *** Fast 4.490 1.159 3.873 0.107723 Dissonant:Fast 0.6338 2.1540 0.294 0.768611

Table C17 Cultural Threat on Musical Parameters

Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 38.3503 2.2837 16.793 5.56e-11 *** Dissonant 13.1036 2.8756 4.557 0.00461 ** Fast_con -2.3171 1.5139 -1.531 0.17764 Dissonant:Fast -0.9253 2.2642 -0.409 0.68286

Table C18 Cultural Competence on Dissonance and Tempo

Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 69.544 2.220 31.329 2.3e-09 *** Dissonant -2.483 1.015 -2.447 0.01689 * Fast 3.046 1.092 2.788 0.00582 ** Dissonant:Fast -1.460 1.988 -0.735 0.46278

Table C18 Cultural Evolvedness on Dissonance and Tempo

Fixed effects: Estimate Std. Error t value Pr(>|t|) (Intercept) 83.6504 1.4823 56.434 < 2e-16 *** Dissonant -2.3876 0.7816 -3.055 0.00253 ** Fast 1.5078 0.7167 2.104 0.03663 * Dissonant:Fast 0.9727 1.4040 0.693 0.48859