Abstract
This study evaluates the influence of cultural and personal aspects of authenticity on the cognitive elaboration of music. Cognitive elaboration of music is comprised of a recipient’s subjective theories about and aesthetical evaluation of (a piece of) music. Subjective theories are based on information from different sources including oneself and the media. They can, for example, be used to explain the liking and perceived aesthetics of a piece of music. We suggest a theoretical model of the positive effects of authenticity on subjective theories and aesthetical evaluation. The model was tested in an online experiment with a randomized 2 × 2 between-subjects design, with cultural authenticity (high vs. low) and personal authenticity (high vs. low) as experimental factors. The levels of authenticity were manipulated in three musical excerpts and musicians’ images induced through album reviews of fictitious bands. Cognitive elaboration was measured by number of ideas related to the music whereas aesthetical evaluation of the music was measured by ratings of liking of the music. A sample of 492 subjects participated. Results indicate a positive effect of cultural authenticity on the function- and affect-based aspect of aesthetical evaluation. The data did not support a positive effect of authenticity on subjective theories.
Keywords
This study addresses the relationship between authenticity in music and the cognitive elaboration of music. Authenticity is a concept discussed by many music fans and scientists in equal measure (see Auslander, 1998; Frith, 2004). Authenticity is often considered an important contributor to the perception and liking of music (see Armstrong, 2004; Dolan, 2010; Edidin, 1998; Frith, 2004). Our goal is to analyze the relationship between authenticity and the evaluation of music using a randomized experiment and to provide support for the validity of this assumption. Kurth (1947) suggests that aesthetical sensations are based on psychological operating conditions. Thus, we define authenticity as such an operating condition that affects the cognitive elaboration of music. Subjective theories of listeners and their aesthetical evaluation of music are defined as two components of this elaboration, in which people engage during their everyday life.
Authenticity
Authenticity in the musicological context carries different definitions depending on the musical genre, the era, and the social reference group. The concept is often discussed in music education (Schippers, 2006) as well as in writings about the aesthetical value of music (Armstrong, 2004; Dolan, 2010; Frith, 2004). Authenticity is often equated with the terms “credibility”, “fidelity”, “traditional and original”, “pure”, “real”, “serious”, “uniqueness”, “historically correct”, or “being true to oneself” (Bennett, 2010; Edidin, 1998; Kahf, 2007; Schippers, 2006).
In the context of music education and world music, authenticity is sometimes addressed as “cultural purity” (Hernandez, 1998; Taylor, 1997). As Redhead and Street (1989) suggest, authenticity can be expressed by the similarity of music to its original source. This can be achieved by using traditional instruments and avoiding modern instrumentations. Music teachers interviewed by Johnson (2000) believe that the use of synthesized instruments, studio production, and modernization reduce authenticity. Other authors propose that commercialization, tourism, and proximity to mass production by the music industry diminish authenticity (Smith & Brett, 1998; Vallely, 2003).
Some authors, on the other hand, use a more flexible approach to define authentic music. They suggest that authenticity may also revolve around an awareness of tradition, which can be achieved by using appropriate costumes and lyrics (Hernandez, 1998; Rees, 1998). All the authors mentioned above use either a culturally- or an ethnographically-based approach to yield examples that comply with their definition of authenticity. However, Matheson (2008) argues that authenticity is achieved by the music’s emotional effect on the listener, which the author terms “emotional authenticity”. Furthermore, Schippers (2006) points out that authenticity can also refer to the strength of the musician’s expression.
Literature regarding classical music uses terms such as “historical correctness” and “Werktreue” (Fábián, 2001, p. 158) to define authentic music. According to these terms, authenticity of a performance depends on how faithful the performance mirrors the intentions of the composer. In historically correct performances, musicians use old instruments and adopt playing techniques which were typical for the era in which the piece of music was composed (Edidin, 1998). However, it is an ongoing discussion whether a piece of music can actually be performed historically correct, as Taruskin (1995) put it: “We cannot know intentions … or rather, we cannot know we know them” (p. 97). Kivy (1995) and Edidin (1998) argue that historical correctness does not ensure the quality of a performance on its own. Therefore, Kivy (1995) suggests to define authenticity differently, namely as personal authenticity. Through this definition, one addresses the artistic aspect of the musician and the importance of individuality and originality, as opposed to historical correctness. Nevertheless, historical authenticity can be the source of aesthetical quality, as both historical correctness and personal authenticity may simultaneously influence the perceived authenticity of a given musical piece (Edidin, 1998).
The two aspects of authenticity, the historical or cultural approach and the personal approach, are also applied to more contemporary musical genres. Since the early 1960s, rock bands such as the Rolling Stones or the Beatles claimed their authenticity by imitating the early Rock ‘n’ Roll- and Blues-sound (Butler, 2003). In the late 1960s, rock musicians and rock aesthetics began to emphasize the importance of individual expression of the musicians. Now, being authentic means being true to one’s own individual artistic visions despite social and commercial pressure (Weinstein, 1998, as cited in Butler, 2003). In progressive rock music, authenticity can be defined by the seriousness of the music and the artistic virtuosity of the musician amongst other things (Sheinbaum, 2008). Furthermore, authenticity can be used to distinguish, for example, rock music, indie music, or hip-hop from mainstream or pop music, which is often considered less authentic (Dolan, 2010; McLeod, 2001; Schippers, 2006). Other important characteristics of authentic hip-hop music are individuality, uniqueness, and the musicians saying the truth as it is (Kahf, 2007; McLeod, 1999).
In pop music, authenticity is not based on whether a piece of music is honest or true in regard to the artist or musical idiom. In fact, pop music itself constructs a sense of realness and thus defines its own aesthetical value (Frith, 2004). Authenticity is often claimed by pop groups using the identity of listeners and by using musicians’ images (Leach, 2001). An example of how authenticity construction can be used for commercial success is the phenomenon of idol-shows. The concept of these shows is based on the public process of creating pop stars (Fairchild, 2004). However, criticism arises that the contestants are not real musicians due to commercialism and manipulation of the competition (Fairchild, 2004). Nevertheless, this highly commercial process can produce authentic contestants, because the publicness of the show is used to construct authenticity. With carefully edited auditions, commentaries, and showcases of big emotions by participants and judges alike, the show points to the seriousness of the contest and those involved. The show is seemingly based on public participation and confirmation. The contestants are presented as not yet “tainted” by marketing, and instead as likable and aspiring; they therefore can be perceived as authentic (Fairchild, 2004).
Following this notion, we can establish a more psychological perspective on authenticity: it is a construct shaped by the recipient in the process of listening (Moore, 2002). However, to the listener, authenticity seems like an objective feature of the music: music fans often discuss authenticity as an essence within music, which can be present or absent (Auslander, 1998). Authenticity can be found in evaluation categories, shown by Appen and Doehring (2000). Different from our definition of authenticity, Appen and Doehring (2000) use the concept, in their evaluation of music critiques, only to distinguish whether a musical album contains truthful emotion and artistic ambition of the artists. The authors suggest that authenticity can be seen as an entrance requirement for positive aesthetical evaluation and liking of a piece of music. An album which is perceived as non-authentic is excluded from a following analysis and positive value judgment. In this study, authenticity is defined as an entrance requirement for subsequent cognitive processes such as the generation of information and subjective theories.
The proposition of authenticity as shaped in the process of listening suggests that it can be manipulated by certain primes and cues, so that any music can be seen as authentic under certain conditions. Based on this proposition, we conceive the terms cultural authenticity and personal authenticity, introduced earlier, as two psychological dimensions of authenticity. That is, cultural authenticity addresses the question whether a composition is perceived by the recipient as being played true to its origin or if it seems to be representative for a certain culture. Culturally non-authentic music appears to be influenced by modernity, for example, through the use of synthesized instruments. Personal authenticity addresses the perceived relationship between a composition and the composer in the sense of honesty, individuality, and independence. Recipients perceive personal authenticity by the musician, if the musician stays true to his or her artistic ideal and by the music’s originality and emotional expression. Personally non-authentic music appears to be played solely to achieve commercial success and lacks individuality. These two types of authenticity also tackle the aspects of self-realization and belonging in the listener as suggested by Weisethaunet and Lindberg (2010). They suggest that cultural authenticity addresses the listener’s need to belong, in that the music belongs to a certain culture or community. The personal authenticity addresses the listener’s need for self-realization, since the music stands for individuality and uniqueness. Note that the proposed dimensions neither contain every possible aspect of authenticity nor do they fully apply to all existent musical genres, such as trance or techno. Nonetheless, the dimensions should apply across most genres. In our experiment, we will manipulate authenticity by using statements found across the music literature. A summary of these is listed in Table 1.
Descriptions of authenticity in the music literature.
Cognitive elaboration of music
In this study, cognitive elaboration of music is defined by two components, subjective theories and aesthetical evaluation of music. We talk and think about music while we use information from different sources, for example, the media, friends, or personal experiences. Based on this information, we generate our own subjective theories to explain our musical experiences. Especially in the case of adolescents, Behne (1987) suggests that they behave like everyday psychologists. Our subjective theories are developed from the everyday search for world- and self-realization (Groeben & Scheele, 2010). They are based on complex reflexive cognition and are maintained to give explanation to the world and to develop prognoses and decisions for action (Groeben, Wahl, Schlee, & Scheele, 1988). The subjective theories of a listener partly determine how he or she constructs music and communicates music with the outside world. Therefore, subjective theories structure the cognitive elaboration of music, in which the listener enriches music with information. Subjective theories can address the music itself, the musician or composer, for example, through knowledge of a certain musical style or a musician’s individual style (Hargreaves & North, 1999). Subjective theories can also address the effect of music on the listener or other listeners (Behne, 1987; Zillmann & Gan, 1997). Moreover, listeners ascribe different functions to the music they like to hear, such as stress relief or social contact (Kleinen, 1986).
We postulate an impact of the two aspects of authenticity on this construction of subjective theories; that is, listeners generate more subjective theories if the music is personally authentic than if it is culturally authentic. Regarding cultural authenticity, music is measured by a certain cultural standard and one can relate subjective theories to the intention of the composer, to a certain cultural background or to a certain way music has been performed during a certain timeframe. Personal authenticity, however, addresses the relationship between musician and music. The variable of the musician as a person is much more flexible than the variable of a cultural standard. Thus, we propose that the larger flexibility leaves more space for the subjective generation of information and the diversity of thought about the music. Therefore, personal authenticity, when compared with cultural authenticity, should lead to more diversity of thought about the music and thus more subjective theories. Because subjective theories are based on information from very different sources, we suggest the idea count, as in the number of ideas a listener generates for a piece of music, as a tentative measure of subjective theories. These ideas represent the information the listener generates and uses for further analyses to eventually form the basic elements of his or her subjective theories.
Aesthetical evaluation of music can be communicated through both conscious verbal actions and through subconscious tradeoff actions such as monetary or time-based investments (Behne, 1987). We define four factors to measure the aesthetical evaluation of music: three factors are defined by Hargreaves, North, and Tarrant (2006) to explain the formation of taste – the characteristics of the music, listener, and context – and the fourth factor is the perceived function of music (Frith, 2004; Kleinen, 1986). Characteristics of music can mean style or complexity, while characteristics of the listener can be age, gender, personality, or musical knowledge (Hargreaves et al., 2006). Characteristics of the context can refer to social situations, such as being by oneself or in a group of people, a certain time-frame in history or a certain country (Hargreaves et al., 2006). The perceived function of music can be one of, for example, distinction or escapism (Dollase, 1997).
Theoretical model
We propose a model of the relationship between perceived authenticity and cognitive elaboration of music (Figure 1). We hypothesize that perceived authenticity is a requirement for further cognitive elaboration of music. Once authenticity is ensured, the music seems worthy of engaging in it. On the other hand, a non-authentic musical piece loses its value for the recipient to engage in further analysis.

Suggested model of the relationship between perceived authenticity and cognitive elaboration of music. Circles stand for dependent and independent variables. Rectangles name additional influences on these variables, which are, however, not tested in the present study.
Music fans differentiate between authentic and non-authentic music (Auslander, 1998). Listeners base their evaluation of the authentic quality of a piece of music upon the music itself but also upon extra-musical beliefs and attitudes (Frith, 1996, as cited in Auslander, 1998). As outlined above and shown in Figure 1, authenticity is constructed through different sources of information and the perception of authenticity can, amongst other things, be actively influenced by certain musicians’ images (Johnson, 2000; S. Jones, 1999). A musicians’ image is often used for marketing purposes and can contain information about, for example, the musicians’ personality (Borgstedt, 2008). The recipient then gives meaning to the image and enhances it with associations and evaluations (Reynolds & Gutman, 1984, as cited in Cohrdes, 2013). For example, a musicians’ image can suggest “honesty” or “realness” of an artist (personal authenticity). It could also describe the musician as culturally authentic, faithful to the origins of the music. Musicians themselves can influence the perception of authenticity by endorsing their cultural background (cultural authenticity). Furthermore, musical elements can suggest authenticity, for example, through a complex played bass line or through a unique voice (personal authenticity).
Our model also addresses the sources of subjective theories, which, as discussed earlier, lie in the utterances or opinions of others or are based on one’s own experiences. Influences on the aesthetical evaluation of music can be found in musical and personal characteristics, in the social context (Hargreaves et al., 2006) or in social functions of music (Frith, 2004; Kleinen, 1986). We propose that authenticity can increase the generation of subjective theories and positive aesthetical evaluation of music. The bold arrows suggest a stronger influence of personal authenticity than cultural authenticity on cognitive elaboration of music (see Figure 1).
Based on this model we propose the following hypotheses:
Effect of authenticity on subjective theories: A higher authenticity in a piece of music leads to the generation of more subjective theories about the music.
Effect of authenticity on aesthetical evaluation: A higher authenticity of a piece of music leads to a better aesthetical evaluation of the music.
Effect of personal vs. cultural authenticity on subjective theories: Personal authenticity of the performer increases the generation of subjective theories more than cultural authenticity.
Effect of personal vs. cultural authenticity on aesthetical evaluation: Personal authenticity of the performer leads to better aesthetical evaluation of music than cultural authenticity.
Method
Overview of the study
We conducted an online experiment with four experimental conditions. Regarding the experiment, we have followed the ethical principles of the German Psychological Society (DGP) and the Association of German Professional Psychologists (BDP). The experimental design was a 2 × 2 between-subjects design with two independent variables with two levels each: cultural authenticity (high vs. low) and personal authenticity (high vs. low). In condition +C+P, both cultural authenticity and personal authenticity are high. In condition +C−P, cultural authenticity is high, whereas personal authenticity is low. In condition −C+P, cultural authenticity is low, whereas personal authenticity is high. In condition −C−P, both cultural authenticity and personal authenticity are low.
The four theoretical hypotheses translate into three contrast hypotheses for both dependent variables, subjective theories, and aesthetical evaluation, respectively. Hypotheses (1) and (2) are tested with the contrasts ψ1 and ψ2. Hypotheses (3) and (4) are tested with contrast ψ3. According to hypotheses (1) and (2), high cultural authenticity leads to a higher count of subjective theories as well as more positive aesthetical evaluation than low cultural authenticity. Contrasts are defined in terms of expected means μ and the hypothesis is:
The corresponding null hypothesis tested is H0: ψ1 ⩽ 0.
The same accounts for personal authenticity: high personal authenticity leads to a higher count of subjective theories as well as more positive aesthetical evaluation than low personal authenticity. The contrast hypothesis is defined as:
The corresponding null hypothesis tested here is H0: ψ2 ⩽ 0.
At the same time hypotheses (3) and (4) state that personal authenticity has a stronger effect on the dependent variables than cultural authenticity, defined in terms of expected means as:
The corresponding null hypothesis tested here is H0: ψ3 ⩽ 0.
Participants
A total of N = 1,251 participants took part in the online experiment. Of these, n = 562 participants completed the study and their data were included in the final analysis. The acquisition and data collection took place from 16 June 2014 until 7 July 2014. Participants were recruited through emails to student representatives, mailing lists of the student councils, Facebook groups, a participant-registration-system of the University of Hamburg (Germany), and various forums. Acquisition took place through correspondence on a total of 15 forums, two black boards, and three Facebook groups. A total of 1479 student bodies from 102 universities were contacted via email. Of these, around 80 email addresses could not be reached. Due to incomplete feedback from the student representatives, an identification of the successfully-reached count of the student body is unknown. Exclusion criteria included proximity to Chinese culture and aged below 18. The participants’ and parents’ place of birth were collected.
Data from 56 participants were excluded prior to final analyses. Among these participants, three were under the age of 18 at the time of their participation, and five, who finished the study, had a China-near cultural background (e.g., Malaysia or Vietnam). In addition, n = 46 participants suggested they were familiar with Chinese music. One participant provided invalid data and another participant explained she had problems with the audio. Thus a total of n = 492 participants (of which n = 310 female, n = 176 male, n = 6 did not specify their sex) were included in the final analyses. The mean age was M = 24.42 years with a range from 18 to 60 years and a median of 23. About 63% stated high school diploma as their highest educational degree, whereas 34% stated a university degree to be their highest educational degree. About 89% of the participants stated they were attending university. Of these students, 47% were studying natural science, 21% humanities, 13% psychology, 8% educational science, 5% economics and social science, 4% medicine, and 1% law. With the use of random assignment, a feature of the online software used for this study, n = 125 were assigned to condition +C+P, n = 124 were assigned to condition +C−P, n = 125 were assigned to condition −C+P, and n = 118 were assigned to condition −C−P. The randomization used a sampling without replacement method, in which the sampling starts anew after every four participant is assigned to one of the conditions.
Materials and procedure
The online experiment was developed using the online software SoSci Survey (Leiner, 2014).
Audio material
In this study, a total of four 10-second extracts of music examples were presented and automatically repeated three times, leading to a 30-second exposure to each music example. The short duration and repetition of each stimulus were applied to limit the cognitive demand for the participants to appreciate the music. As described later in the text, the participants also had to include information on the musicians in their evaluation about the music and the musician. To ensure a standardized stimulus presentation, the web page was forwarded automatically after each 30-second long extract. During the presentation of the music example, participants could not forward the page. The first music example was a piece of Salsa music and was presented without manipulation of its authenticity. 1 The reason for its use is that it made the cover story more plausible while giving the participants the opportunity to practice the task.
For the following three music examples, cultural and personal authenticity were manipulated using musicians’ images. To avoid an already present connotation of the music by the listener, foreign musical pieces in the style of traditional Chinese music were newly composed using the program Reason 6© (Propellerhead). The use of Chinese music could keep the participants’ familiarity with the music at an almost constantly low level. This criterion of unfamiliarity for the western listeners was achieved by using the pentatonic scale as well as traditional Chinese musical instruments. It also enabled us to declare one sound in the examples as either a traditional or a synthesized instrument under the different conditions. For each music example a fictitious band name, album name, and song title were invented (see Appendices A, B and C for the notation of the examples and the Supplemental Material Online section for audio files). 2 The music consisted of traditional Chinese instruments (e.g., Gu Zheng) and synthetic instruments (e.g., electronic bass or electronic drums). We assumed that describing this music in one experimental condition as traditional and in another condition as modernized and synthetic to be reasonable at least for the layperson. The order in which the examples were presented was fixed across all participants. For reasons of simplicity, the Chinese music examples will be referred to as MEA, MEB and MEC. Before and during the presentation of the music examples, the image texts were presented to manipulate the independent variables: cultural and personal authenticity, respectively.
Manipulation of authenticity
For the induction of cultural and personal authenticity, album reviews were written and adapted to each condition. The literature on authenticity in music supplied keywords of high or low authenticity, as described in Table 1. For example, the keyword “perpetuation of tradition” (C. Jones et al., 2005) was used to define high cultural authenticity, whereas the keyword “influenced by modernity” (Johnson, 2000) was used to describe low cultural authenticity. On the other hand, “style and originality” (Kivy, 1995) stands for high personal authenticity and “commercialization” for low personal authenticity. A total of 16 album reviews were created and then evaluated in a preliminary study for their suitability regarding the manipulation of cultural and personal authenticity by n = 10 participants (6 female; Mage = 26.40; SDage = 4.01). The questionnaire for testing the manipulation consisted of 4 items addressing cultural authenticity, 4 items addressing personal authenticity, 2 items addressing a possible positive or negative judgment by the album review, and 1 item to increase attention (see Table S1 in the Supplemental Material Online section). For 12 album reviews, analyses showed that the keywords of high or low authenticity were sufficient to trigger the corresponding subjective impression of high or low authenticity. These reviews served as stimuli in the main experiment (see Table S2 in the Supplemental Material Online section).
Procedure
At the beginning of the experiment, a cover story informed the participants about the study’s purpose. They were told that several studies had been conducted to examine opinions and attitudes of western listeners toward world music from Africa and Latin America in a broad-based project of the fictional database of world music. The present study would be a continuation of the project, in which Asia was being examined via world music from China.
The online questionnaire comprised three work phases. The first work phase initially captured general sociodemographic information. Afterwards, the participants were prompted to activate loudspeakers or put on headphones. By using a short testing music sequence containing simple piano sounds, participants were able to test their audio adjustments. To familiarize the participants with the task, a training block with the non-manipulated music example was presented with the task of generating ideas. For this, it was explained to the participants that a music example from a previous study about world music from Latin America would be presented to them. To make it easier for the participants, they were first shown fictitious examples of ideas from two alleged earlier participants about the Salsa music. Their ideas differentiated in count and complexity to show different possibilities of working on the task. One example participant included 12 ideas with phrases such as, “one could dance easily to this music”, while the other only provided three ideas in short phrases such as “Party and Salsa”. After the presentation of the fictitious ideas and Salsa music, participants were prompted to write down their own ideas to the music. The participants were asked to use all available time and not to abandon their work space. Afterwards, participants were asked about their audio option (e.g., headphones or speakers) and their typing speed, which served as a control variable.
In the second work phase, the experimental music examples were presented and the dependent variables were measured together with the manipulation check for each example, respectively. It was explained to the participants that they will be hearing three Chinese music examples to which they will respond on two tasks. In the first task they would write down ideas and in the second task they would evaluate statements about their perception and their attitude toward the piece of music they have just listened to. The first task measured subjective theories, and the second task measured aesthetical evaluation and comprised the manipulation check. Participants were then informed that they would be presented with album reviews about the bands, which they should read carefully. For all three music examples, the album reviews were presented before and during the music presentation. After each 30-second audio presentation, participants were given the idea generation task followed by the aesthetical evaluation task.
The third work phase consisted of a closing questionnaire. Afterwards, participants could state what they still wanted to convey about the topic or about the study in general. At the end of the experiment, the background and the goal of the study were clarified. In total, the experiment lasted 20–30 minutes. As a reward, a gift card worth 20 euros and three gift cards each worth 10 euros were raffled to the participants. All data was collected anonymously and the email addresses of those participants who wanted to participate in the gift card drawing were recorded separately from the entered data during the online study.
Measures
Measurement of dependent variables
For each of the three Chinese music examples, the two dependent variables, subjective theories and aesthetical evaluation, were measured. Subjective theories was measured by the number of ideas a participant generated for a music example, following the instruction to write as many ideas as possible related to each music example. These ideas could be simple associations or complex ideas about the relationship between the music and the listener, other listeners, and/or the musician. In contrast, participants were asked to refrain from giving evaluations about the musical aesthetics, such as, “I think the music is beautiful”. The participants were given a maximum of four minutes to finish the task. After four minutes the webpage was forwarded automatically. To enforce a certain minimum time-frame for generating ideas, the forward button appeared only after one minute. Due to restrictions from the online software, the maximum idea count was 25.
The second dependent variable, aesthetical evaluation, was measured using three dimensions: music-related liking (4 items), person-related liking (3 items), function and affect (5 items). The items for liking were adapted from Cohrdes, Lehmann, and Kopiez (2011) and also comprised self-developed items. The items for function and affect were constructed according to Kleinen (1986) and Dollase (1997). The items of the present questionnaire were 6-point Likert-scales ranging from 1 = “does not apply at all” to 6 = “applies fully”. Thus the dependent variable aesthetical evaluation is measured by the variables music-related liking, person-related liking, and function and affect. Table 2 contains statistics of the item analysis for the aesthetical evaluation-scale (for the items see Table S1 in the Supplemental Material Online section).
Statistics of the item analysis for the aesthetical evaluation-scale for all three music examples MEA, MEB, and MEC..
Note. For the items, see Table S1 in the Supplemental Material Online section; rit: Item-total correlation; α: Cronbach’s α without this item. Item 3.4 was eliminated in the further analysis and Cronbach’s α for music-related liking increased to α = .85 under MEA, α = .90 under MEB and α = .89 under MEC.
Manipulation check and covariables
The online questionnaire also comprised a manipulation check and the measured covariables. The manipulation check consisted of 3 items relating to subjective cultural and personal authenticity, respectively. Manipulation check items were presented in randomized order and together with the aesthetical evaluation items. Covariables were person-related variables like age and gender. Further covariables were musical experience, interest in music, self-evaluated musical competence (Lehmann, 2011), seriousness of participation, familiarity with traditional Chinese music, general interest for the musician behind the music one listens to, typing speed, and general willingness to write. The general willingness to write was operationalized by counting the ideas (idea count) to a non-manipulated music example, in this case, the Salsa music presented at the beginning of the study. In addition, the openness scale of the NEO-FFI by Costa and McCrae (Borkenau & Ostendorf, 1993) and the short test of music preferences (STOMP; Rentfrow & Gosling, 2003) were used. The internal consistency of all the NEO-FFI-Scales lie between r = .71 and r = .85 (Borkenau & Ostendorf, 1993). With a repetition interval of two years, Borkenau and Ostendorf (1993) report a retest reliability between .65 and .81. Rentfrow and Gosling (2003) report retest reliabilities between .77 and .89 for STOMP.
Results
The participants of our study were highly interested in music: 42% of them chose the highest rating category on the item “Music interests me very much”. Furthermore, the manipulation check proved to be largely successful; that is, both conditions +C+P and +C−P showed higher means on the cultural authenticity scale than the conditions −C+P and −C−P and yielded a significant medium effect (ψ = 0.28, t488 = 4.38, p < .001, d = 0.39). Likewise, both conditions +C+P, −C+P showed higher means on the personal authenticity scale than the conditions +C−P, −C−P and yielded a significant large effect (ψ = 0.60, t488 = 9.99, p < .001, d = 0.9). No significant mean difference was found between conditions +C+P and +C−P on the cultural authenticity scale (ψ = 0.03, t247 = 0.43, p = .670, d = 0.04), but, in contrast, condition −C+P showed a higher mean on the cultural authenticity scale than condition −C−P (ψ = 0.24, t241 = 2.62, p = .009, d = 0.33). No significant mean difference on the personal authenticity scale was found between conditions +C+P and −C+P (ψ = 0.08, t248 = 1.05, p = .294, d = 0.13) nor between conditions +C−P and −C−P (ψ = 0.14, t240 = −1.45, p = .147, d = 0.19). In summary, the manipulation of personal authenticity was largely successful, whereas cultural authenticity was not as unequivocally manipulated according to the manipulation check.
For the test of our statistical hypotheses, the statistical contrasts are tested in a regression analysis while controlling for covariables. Table 3 contains the correlation analyses between tentative covariables and dependent variables. Table 4 contains the descriptive statistics of the dependent variables, idea count (Y1.1) and the three dimensions of aesthetical evaluation (Y2.1 – Y2.3).
Correlation analyses between tentative covariables and dependent variables Y1 subjective theories and Y2 aesthetical evaluation.
Descriptive statistics of the dependent variable Y1 subjective theories and Y2 aesthetical evaluation for all three music examples MEA, MEB, and MEC..
Note. ME = Music Example A to C; M = mean; SD = standard deviation. MSw = mean squares within groups, used to calculate t-values and effect sizes.
Subjective theories
On average, participants named 4.79 ideas per music example with a range from 0 to 20. Contrary to expectations, the mean values of idea count under condition +C+P (see column 1 of Table 4) are below the means of the other three conditions (see columns 3, 5, 7 of Table 4). Furthermore, mean values of idea count under condition +C−P are below those of condition −C+P, which we did not expect. The mean values for MEB and MEC under condition −C−P (see column 7 of Table 4) prove to be the highest among all four conditions.
Contrast hypotheses are tested against H0: ψ ⩽ 0 and with the alternative H1: ψ > 0. For hypotheses (1) and (3), which focus on the dependent variable Y1.1 idea count, Poisson regression is implemented. An α-adjustment is avoided due to the explorative character of this study. Table 5 contains the results of the regression analyses. Results show that the mean differences are contrary to expectations (negative contrast values). Effect sizes for the effect of authenticity on idea count remain small with a range from |reffect size|Min = 0.01 to |reffect size|Max = 0.10. reffect size is calculated by correlating the variables idea count with group membership (lambda weight for the contrast) (Sedlmeier & Renkewitz, 2013).
Multiple regression.
Note. ME = Music Example A to C. Contrast 1 = Comparison of low cultural authenticity vs high cultural authenticity (conditions +C+P and +C–P vs –P+C and –C–P). Contrast 2 = Comparison of low personal authenticity vs high personal authenticity (conditions +C+P and –C+P vs +C–P and –C–P). Contrast 3 = Comparison of the effect of cultural authenticity vs the effect of personal authenticity (condition +C–P vs –C+P). B = beta coefficient; p = p-value. Significant effects are boldface. Pseudo R2 for Poisson regression is calculated using the formula for McFadden’s (1974) pseudo-R2 (1-(ResidualDeviance/NullDeviance)).
Aesthetical evaluation
Regarding aesthetical evaluation, participants rated the music examples negatively. With a scale center at value 3.5, mean values are at 2.86 on the music-related dimension, 2.48 on the person-related dimension, and 3.59 on the function- and affect-dimension.
Contrast hypotheses are tested against H0: ψ ⩽ 0 and with the alternative H1: ψ > 0. For the regression analysis regarding hypotheses (2) and (4), which focus on the dependent variables Y2.1 – Y2.3 for aesthetical evaluation, a linear regression is used. Results show small effect sizes for music-related liking, ranging from dMin < 0.01 to dMax = 0.11 (see Table 5). For the person-related dimension, effects range from dMin < 0.01 to the significant effect dMax = 0.15 for contrast ψ2 under MEC. Results for the function- and affect-dimension show effect sizes range between dMin < 0.01 to dMax = 0.17. Herein, contrast ψ1 is significant for MEA (d = 0.17) and MEC (d = 0.16).
Covariables
Furthermore, the general willingness to write is significantly associated with the idea count under all three music examples (MEA: B = 0.08, z = 15.70, p < .001; MEB: B = 0.08, z = 14.02, p < .001; MEC: B = 0.08, z = 13.40, p < .001). The variable openness has a significant effect on the idea count only under MEC (B = 0.08, z = 2.17, p = .030). Of the three aesthetical evaluation dimensions, only the function and affect variable has a significant effect on the idea count under the first two music examples (MEA: B = 0.08, z = 2.40, p = .016; MEB: B = 0.09, z = 2.67, p = .008). In addition, openness and familiarity with Chinese music have a significant effect on all three dimensions of aesthetical evaluation under all three music examples. Finally, the general interest in musicians the participants listen to was associated with person-related liking (MEA: B = 0.13, t484 = 3.55, p < .001; MEB: B = 0.17, t484 = 4.41, p < .001; MEC: B = 0.13, t484 = 3.49, p < .001). Figure 2 depicts estimated means and standard errors of means of the dependent variables for each music example.

Means and error bars (standard error of mean) of AV Y1 idea count and AV Y2 aesthetical evaluation for each music example. The three tick marks on the right of each graph represent the grand mean ± MSw1/2 / 2. The effect size d between the means can be read in comparison with this axis.
Discussion
Though carefully planned as a randomized experiment and using important covariables, the study yielded non-significant effects of authenticity on the generation of subjective theories and aesthetical evaluation. Thus, the predicted effects of both aspects of authenticity stated in hypotheses (1) and (2) and which predominate the literature, was not found. Moreover, hypotheses (3) and (4) stating that personal authenticity has a greater effect than cultural authenticity were found to be non-significant and, moreover, corresponding mean differences were contrary to expectations. However, regarding hypothesis (2), an effect of cultural authenticity on one of the three dimensions of aesthetical evaluation was found: the function- and affect-based dimension. That is, traditional authentic and non-modern descriptions of music lead to better function- and affect-based liking. The listeners seem to ascribe such music more positive emotional effect and usage. The finding supports Frith’s (2004) assumption that authenticity is largely associated with the social functions of music, though just for the cultural part of authenticity. Zanes (1999, as cited by Weisethaunet & Lindberg, 2010) also believes that there is a strong relationship between authenticity and the construction of identity, it being one of the functions ascribed to music. Apart from this, another significant effect regarding hypothesis (2) was found for personal authenticity, which pertains to the person-related liking. This suggests that participants in the condition of high personal authenticity indicate greater interest in the musicians and a desire to know more about them.
The non-significant findings do not support the theoretical hypotheses carefully derived from the literature. Many authors predict a strong influence of authenticity on the perception and liking of music. Appen and Doehring (2000) understand authenticity as a prerequisite for positive evaluation of music. Based on this proposal, we predicted that authenticity should enhance the generation of ideas about the music. In contrast, the non-significant findings do not support the assumption that perception of authenticity works as an entrance requirement for further cognitive elaboration of music. In fact, the effect appears to be in the opposite direction. Furthermore, we could not find a stronger influence of personal authenticity on the generation of ideas about the music than that of cultural authenticity. A tendency to the contrary effect appears to be the case. Relating to aesthetics, Edidin (1998) assumed that cultural as well as personal authenticities contribute to the aesthetical judgment of music. Our results suggest that personal authenticity tends to contribute more to a person-related liking of the music, whereas cultural authenticity is more associated with a function- and affect-based liking of the music. Nonetheless, the lack of significant findings suggests that the influence of authenticity on aesthetical evaluation of music is highly questionable. Moreover, we could not find a greater influence of personal authenticity on the aesthetical evaluation of music than cultural authenticity. Thus, the assumption of Kivy (1995) and Edidin (1998) that personal authenticity rather than cultural authenticity is decisive for aesthetical evaluation finds no support in the data. A tendency to the contrary effect was found mainly for the function- and affect-based liking of music.
After all, which predictor in the design accounts for the generation of subjective theories? The generation of subjective theories seems to be associated with the aesthetical evaluation and liking of a piece of music rather than with authenticity. The data shows that the generation of ideas correlates with function- and affect-based liking. This highlights the importance of the attributed functions and the affect-laden perception of a piece of music in the generation of ideas. Idea count seems to be a valid representation of the information listeners generate, which in turn, as discussed above, helps formulate subjective theories. However, relating to the methodological approach for the measurement of subjective theories, idea count may not represent the best empirical variable. Not all ideas are subjective theories, so we need further qualitative evaluation of ideas and how they relate to subjective theories.
Results of correlative tests show that interest in the artist is associated with person-related liking with a middle-sized effect (rY2.2 = .21). Thus, a participant who generally shows more interest in the artist who presents the music will evaluate the musician of new and foreign music more positively. Openness also seems to play a significant role in the generation of ideas as well as in the aesthetical evaluation of music. The effects remain small- to middle-sized (rY1 = .15, rY2.1 = .30, rY2.2 = .32, rY2.3 = .26), indicating that a personality with a high value on the openness scale would be expected to generate more ideas and evaluate the music pieces more positively.
The analyses show that the test sample consists mostly of those who are very interested in music. As the participants were largely students, the generalization of the results is limited to a certain age range and level of education. However, the attempt to draw a sample of students from a wide range of fields of study was largely successful. Due to the nature of the study as an online experiment, the reasons for the abortions could not be controlled or monitored. Due to this unknown self-selection as well as the selection by adopting an online study design, any interpretation of the results is restricted to an internet-using sub-population willing to take part in this kind of experiment.
A noteworthy feature of the study is the quality of the authenticity manipulation. The manipulation has been validated in three ways. First, regarding the content validity, the manipulation was carefully established based on the literature. Second, a preliminary study was conducted for the purpose of validating the authenticity manipulation. Third, a manipulation check was implemented in the main experiment. However, it could not be ensured that the image texts have exclusively manipulated the authenticity and not also positive or negative judgment of the musicians. Participants in the preliminary study indicated that the image texts did contain evaluative characteristics. According to the results of the preliminary study, a description of the band as culturally authentic and personally authentic seemed to contain the most positive judgment, while musicians described as non-authentic seemed to be negatively judged by the image texts. A band described as personally but not culturally authentic seemed to be more positively evaluated by the image text than a band described as culturally but not personally authentic. Thus, the independent manipulation of authenticity without manipulation of judgment of the musicians remains questionable. As described in the theoretical part of this paper, many authors assume that authenticity in itself implies a valuation. For instance, Schippers (2006) defines the construct authenticity as inherently value-based. The term authenticity is also often used in everyday speech, regarding musical themes, for upvaluation and devaluation (see Widdicombe & Wooffitt, 1990). The different level of liking due to the manipulation of authenticity through the album reviews may have influenced the results. Thus, further research should focus on the disentangling of authenticity and liking.
Another notable feature of this study is the usage of foreign instrumental music. All three music examples were without voice. This might have diminished the effect of personal authenticity in particular, since personal authenticity deals with the relationship between the music and the musician. This relationship can be better and more intuitively perceived by the voice of a singer, particularly if a voice is unique and charged with emotions (Appen & Doehring, 2000). In our study, the musicians and personalities were obscured and could not be perceived through the musical stimulus, so they could not use a unique voice to claim personal authenticity. The musicians were only perceivable through instruments, playing techniques, and thus musically subtle nuances, which could be difficult for a layperson to appreciate on first hearing.
Another limitation focuses on the cultural authenticity. The information given in the album reviews could have highlighted the differing cultural background between the listener and the music. Thus, music from one’s own cultural background may lead to different results. This should be analyzed in future studies.
Analyses of the collected ideas led to some interesting findings. First, the collected ideas show different opinions towards cultural authenticity regarding the same music example, independent of condition. For example, some participants found a piece of music, which was described as highly culturally authentic, rather modern. In contrast, other participants in a low cultural authenticity condition found the same piece of music traditional. Second, types of ideas varied clearly between participants. Some gave only simple associations, while others described theories and wrote in complex contexts. Third, collected ideas include theories addressing the topics described earlier: the music itself, the musician or composer, and the affect and function of music. Regarding these varying responses of participants, further research could evaluate the differences and analyze what types of ideas are generated under what condition. One could examine if participants who are confronted with culturally authentic music are more inclined to generate ideas about the music itself, while participants confronted with personal authenticity are more likely to generate ideas related to the musicians.
In conclusion, there does not seem to be a significant relationship between authenticity and subjective theories in regard to the proposed model and in the sub-population considered. A tendency contrary to what can be found in the musicological literature seems to hold: a higher authenticity seems to affect the generation of ideas negatively.
Footnotes
Appendix A
Appendix B
Appendix C
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
