If it’s Mozart,it must be good? The influence of textual information and age on musical appreciation

Abstract

In a framing experiment, 170 participants aged 19–80 years were asked to read a description in the fashion of a program note prior to listening (individually via headphones) to a sinfonia by Josef Mysliveček (1737–1781). Divergent versions of this description were created for treatment manipulation, while the participants were not informed about it. Within a 2 × 2 design the descriptions (a) attributed the musical piece to different composers of highly different prominence and prestige. Half of the participant group was informed that they would be listening to the overture to the pastoral opera Ascanio in Alba by Wolfgang Amadé Mozart (1756–1791), whereas the other half was informed correctly. The composers’ names were (b) combined with descriptions that applied either an analytic or expressive writing mode. Subsequently collected ratings for liking and a number of perceived musical characteristics were significantly higher when participants had read the expressive compared with the analytic writing mode. Interestingly, younger adults showed higher liking ratings when the music was attributed to Mozart, whereas no significant differences were found in older adults. In sum, this study supports the notion that being exposed to text information prior to listening to music affects perception and appreciation of musical characteristics.

Keywords

experimental aesthetics listening music appreciation older adults prestige effect program notes social psychology

Seeking to acquire knowledge or consulting information about the music heard before or after listening to it is a widespread phenomenon especially related to the engagement with so-called western classical music. Several types of information storage are at the recipient’s disposal, usually dependent upon the situation in which the music is heard: a CD is usually accompanied by a booklet, the radio host comments on the musical pieces played, and in concerts, it is common to provide program notes. Among them, the program note appears to be most closely related to the actual listening experience, not least because the original purpose of program notes was/is to advance the recipient’s knowledge and understanding of the music to be heard in order to provide a proper preparation for the concert (Bashford, 2003; Tewinkel, 2016; Waxman, 2012). Typically, program notes contain contextual information about the composer’s biography and the context of the composition of the pieces and provide insights on the interpretation of the music heard, such as information on its structure, meaning, or aesthetic value. As the musicologist Christian Thorau (2013) points out, the verbal information establishes orientation knowledge for the recipient that prepares and accompanies the experience of the object in question. Thereby, verbal information very likely influences the quality of the listening experience and leads to a commented mode of perception that depends on form and content of the knowledge acquired.

In support of this notion and in line with priming studies in other perceptual fields, a number of experimental studies have demonstrated that manipulations of prior information about the music to be heard (or played) can affect listeners’ appreciation of the music (see Hargreaves, 1986, pp. 194–198 for a review; Crozier and Chapman, 1981, for a review across all art forms). Additional evidence for effects of prior information was also provided by a number of more recent studies on the perception, enjoyment, and evaluation of music (Anglada-Tort & Müllensiefen, 2017; Kroger & Margulis, 2016; Margulis, 2010; Margulis, Kisida, & Greene, 2015; North & Hargreaves, 2005; Vuoskoski & Eerola, 2015; Ziv & Moran, 2006), art works, such as digital reproductions of paintings (Kirk, Skov, Hulme, Christensen, & Zeki, 2009; Lauring et al., 2016; Lin & Yao, 2017), photos (Tousignant & Bodner, 2014), poems (Blohm, Menninghaus, & Schlesewsky, 2017), or live theater performances (Wagner et al., 2015).

Effects of prior information have been discussed as being the specific result of framing, which means that something is contextually framed by certain information. More specifically, framing effects are assumed to occur when contextual information, such as different descriptions, attributions or characterizations of a subject matter, leads to systematically different results regarding the audience’s perception and evaluation of stimuli. In their seminal work, Kahneman and Tversky (1979, 1984) have demonstrated that individuals’ interpretations and judgments as well as inference making can be considerably influenced by how stimuli are contextualized or framed.

Framing by written contextual information

As noted by Entman (2003), framing essentially means highlighting selected parts of information inherent to an object or stimulus, making it more salient to the audience in such a way as to “promote a particular interpretation, evaluation, and/or solution” (Entman, 2003, p. 417; see also Entman, 1993; Pan & Kosicki, 1993). Moreover, framing may have a particularly strong potential for influencing the interpretation and evaluation of a subject matter if the words and images that are used in the framing information are “emotionally charged” (Entman, 2003, p. 417). Along these lines, Vuoskoski and Eerola (2015) investigated whether different descriptions either comprising sad or neutral narratives contribute to emotions induced by music. Results showed that musically induced sadness can be intensified by reading a sad narrative description prior to music listening.

Furthermore, the listeners’ attention and enjoyment are seemingly capable of being influenced by written information. Recent research on the influence of program notes indicated that explicit information may elevate listeners’ (children’s) attention as well as their comprehension of the music (Margulis et al., 2015). Another study revealed that short music descriptions (being either dramatic or structural) resulted in less enjoyment of the music compared with listening without reading any texts prior to listening (Margulis, 2010). These results match those observed in a recent qualitative study on audience reactions to program notes by Bennett and Ginsborg (2018), who observed that only 39% of the listeners reported a positive impact of program notes on their listening experience.

Prior information may not just change the listeners’ emotions, attention, or enjoyment of the music, but might also affect more fundamental aspects of the way the music is perceived (Crozier & Chapman, 1981; see also North & Hargreaves, 2008). The perception of basic musical characteristics such as dynamics or affective character of a piece of music may be strongly influenced by prior information. Supporting this notion, research by Chapman and Williams (1976) indicated that the context in which music is heard does not merely induce effects in evaluations but also operates at higher levels, causing changes in subjective listening experiences.

Writing modes

The studies mentioned above have clearly demonstrated effects of information on music perception. However, none of these studies has used a theoretical framework guiding structured manipulation of information in terms of writing modes. The philosopher Albrecht Wellmer (2009) conceptualized the practice of speaking and writing about music, or more generally, verbal notions about music as an interfering with the music itself, and as a consequence, as an interfering with the listener’s experience. Wellmer differentiates between three main dimensions of a decoding listening (“verstehendes Hören”)¹ that developed its respective way of speaking and writing about music: listening to music as (1) sound qualities and their variation, (2) structural development, and (3) meaning and significance (“Bedeutsamkeit,” see Wellmer, 2009, pp. 108–109).

Each dimension frames the music heard differently and, accordingly, each dimension centers a particular focal point within the music heard. According to the theoretical frame of Wellmer’s dimensions of decoding listening, we assumed that each way of speaking and writing about music—below referred to as “writing mode”—can be considered as an interfering factor with regard to the recipient’s appreciation of a wide range of musical characteristics. This implies that the quality of touristic listening (Thorau, 2013; Thorau, in press) initiated by reading program notes prior to the listening experience is highly dependent on the writing modes occurring in program notes. However, there have been no controlled studies that compare the influence of diverging descriptions applying distinct writing modes on the appreciation of musical characteristics in more detail.

Effects of prestige

Another way of manipulating prior information is termed “attribute framing” (Levin, Schneider, & Gaeth, 1998), involving divergent qualities (e.g., positive versus negative connotations) attributed to the musical piece in question. In a recent study, Anglada-Tort and Müllensiefen (2017) investigated the differential impact of varying information on liking of the music heard: recordings that were attributed to a high prestige of the performers received significantly higher evaluation ratings than recordings that were attributed to medium or low prestige. North and Hargreaves characterized this kind of informational influence in the following illustrative way: “if someone is told that a piece of music they don’t know is by Mozart then they will probably like it more than if they are told it is by an undistinguished composer” (North & Hargreaves, 2008, p. 97). Here, the attribution to Mozart is assumed to have an effect of prestige on the evaluative responses due to the name recognition of Mozart, his positive image as a prodigy and genius composer of high prestigious status—true to the motto: “If it’s Mozart, it must be good!”

Indeed, a corpus study on musical attribution which investigated 168 different kinds of text (e.g., concert reviews and program notes) suggested that the “overall evaluations” of the quality of a Sinfonia Concertante for Winds were more positive by music critics who believed the musical work to be attributed to Wolfgang Amadé Mozart (1756–1791) compared with cases when it was thought to be by an unknown composer (Spitzer, 1987). However, the question whether prestige suggestion would also affect listeners’ evaluative judgments if a piece of music would be (1) attributed to Mozart or (2) to an unknown composer has not yet been empirically answered.

Individual differences

Moreover, little is known about these effects as a function of individual differences such as listening experience and suggestibility in particular. For instance, older concertgoers who are routinely involved in music and have thus acquired a high amount of diverse musical knowledge over years may be less susceptible to prior information, that is, they would be expected to show a higher degree of field independence (Crozier & Chapman, 1981; van Paasschen, Bacci, & Melcher, 2015) than younger and less experienced listeners (for a review on age differences in suggestibility see Umanath, 2016). To our knowledge, there has been no attempt to investigate the impact of framing, including prestige suggestion, on the perception and appreciation of music in older adults, since all previous studies have been conducted using samples of children, college students, or young university students.

In the present study, we conducted a listening experiment applying the framing paradigm to investigate whether and how the perception and appreciation of a piece of music might be influenced by information provided similarly to descriptions in a program note. Prior to listening to music individually via headphones, we exposed participants to texts which we manipulated on two dimensions: (1) writing mode of the description (analytic versus expressive), and (2) authorship (attribution of the musical piece to two different composers of highly different prestige). To investigate effects of manipulation on these dimensions, we applied a between-subject design with random allocation of participants to specific combinations of writing mode and authorship.

With regard to the first type of framing information (writing mode), we hypothesized, first, that the different writing modes would have an effect on (a) liking of the musical piece as a whole and (b) the appreciation of musical characteristics inherent to the music. Regarding the second type of framing information (authorship), we hypothesized that liking ratings would be higher for music that was attributed to Mozart and that the evaluation of musical characteristics would differ depending on the authorship information. In addition, we explored if such effects might be moderated by age and musical expertise.

Method

Participants

A total of 170 participants (56% female, M_age = 45.8 years, SD_age = 17.3, age range: 19–80 years; see histogram of age distribution in the online supplemental material) were recruited by announcements via a range of different cultural online forums, a local radio broadcasting (hr2-Kultur) and flyers with the headline “Golden ears wanted!”. Recruitment was conducted under the pretext of a cover story, telling that the study would be about the investigation of historically informed performance practice (e.g., asking questions about the subtle tonal influence of historical musical instruments or playing techniques on the evaluation of compositions). Inclusion criteria were: minimum age of 18 years, excellent knowledge of the German language, and having no hearing impairments. Participants were from the region of Frankfurt am Main (Germany). Most of them can be described as music enthusiasts or music lovers who are interested in classical music and are regular concertgoers. Participants’ mean score on the Gold-MSI scale General Musical Sophistication (Müllensiefen, Gingras, Musil, & Stewart, 2014) was 82.7 (SD = 18.2), which is higher than the 70th percentile of a German sample (Schaal, Bauer, & Müllensiefen, 2014).

All participants received €10 for their participation in this experiment, which was accompanied by a second experiment focusing on the influence of textual information on the evaluation of different musical interpretations (not reported here). Data collection took place at the Max Planck Institute for Empirical Aesthetics in Frankfurt am Main from December 2015 to February 2016. All participants gave written informed consent. The study was approved by the Ethics Council of the Max Planck Society.

Research design

We implemented a 2 × 2 between-subjects factorial design with factors composer and writing mode fully crossed. Prior to listening to music, each participant was requested to read a text presenting information about composer and composition. The texts presented information on Mysliveček or Mozart (factor composer), combined with descriptions of the composition that was either written in an analytic or expressive mode (factor writing mode).

Participants were randomly allocated to conditions, with n = 43 participants in each of the Mysliveček + analytic and the Mysliveček + expressive conditions, and n = 42 in each of the Mozart + analytic and the Mozart + expressive conditions. One-way ANOVAs by condition found no evidence for significant differences between groups in age, sex, education, or scores on the Gold-MSI scale General Musical Sophistication (ps > .63).

Materials

Music

All participants listened to the same widely unknown musical piece by Josef Mysliveček (1737–1781): Sinfonia in E-flat major for 2 horns, 2 oboes and strings, 1st movement “Spirituosa,” from a recording with the music ensemble Concerto Köln and the conductor Werner Ehrhardt (ARCHIV Produktion: 2006, Track 13, duration 3:37 minutes).

Texts

Each text consisted of two sections, one introducing the composer, and the other describing characteristics of the piece of music presented to participants afterwards. For each of the two factors composer and writing mode, two conditions were generated.

Factor composer

Texts about the composers Mysliveček or Mozart included aspects such as family background, reputation, and contextualization of the work – the Sinfonia in E-flat major or the alleged overture to Ascanio in Alba – within the oeuvre. This section was written in accordance with the current state of musicological research about these two composers.

Factor writing mode

Characteristics of the piece of music were described in an analytic or expressive writing mode. The analytic writing mode used a rather objective, analytical language known from the established vocabulary of music analysis. This writing mode is related to the second dimension of decoding listening, namely, structural development (see Wellmer, 2009). The second type of description was written in an expressive writing mode, which is associated with the third dimension of decoding listening (Wellmer, 2009). This dimension, namely, listening to music as meaning and significance, offers uncountable options in terms of its verbal articulation. In order to transfer the dimension into the experimental setting, we chose to define an expressive writing mode as focusing on the affective quality of the music described.

Full texts for each condition can be found in the online supplemental material.

Measures

Liking

Individual positive evaluation was measured by a single item asking “How did you like the movement?” (“Wie gut hat Ihnen der Satz gefallen?”) with a response scale ranging from 1 (not at all/überhaupt nicht gefallen) to 6 (very much/sehr gut gefallen).

Perceived musical characteristics

To assess perceived musical characteristics, participants responded to 28 semantic differentials presented in randomized order with a scale range from 1 to 7 querying aspects of musical structure, sound qualities, expressivity, originality, and affective character (see full list of items in the online supplemental material). To reduce the information to broader dimensions of perceived musical characteristics indicated by homogenous item sets, we subjected the 28 differentials to an exploratory factor analysis with oblique promax rotation. Parallel analysis suggested extraction of four factors (Eigenvalues: 9.0, 2.1, 1.8, 1.2), and rotation yielded clear loading patterns of items on factors with the exception of item 17 (Fluent–Stagnant) which had equally sized loadings on two factors and was therefore discarded (see online supplemental material for loadings, a parallel analysis plot, and item-scale key). The factors were named Quality of Artistic Expressivity (QoAE; 11 items, α = .89), Affective Character (6 items, α = .69), Vitality (6 items, α = .78) and Emotional Expressivity (4 items, α = .85). The first factor was named Quality of Artistic Expressivity due to the highly evaluative character of the differentials such as Pleasant–Unpleasant or Interesting–Boring. The second factor (Affective Character) comprised pairs of adjectives that are metaphorical in character, such as Light–Heavy or Hard–Soft. The third factor included pairs of adjectives (e.g., Static–Dynamic or Slow–Fast) that represent aspects of musical vigour; therefore, this factor was named Vitality. The differentials (e.g., Warm–Cold; Agitated–Calm) of the fourth factor (Emotional Expressivity) refer to expressed emotional facets of the music heard. Mean scores were generated for each factor. Correlations between factor scores were in the range of .35 < r < .63. Correlations of factor scores with liking ratings were in the range of .38 < r < .70.

Level of familiarity with composer and music

To check for familiarity with composer and music listened to, we asked participants if they were familiar with the composers’ music in general (“Have you heard a composition by Mysliveček [Mozart] before?”/ “Haben Sie schon Kompositionen von Mysliveček [Mozart] gehört?”) or with the specific movement just heard (“Were you familiar with this movement by Mysliveček [Mozart]”/ “Kannten Sie bereits den Satz aus der Sinfonia von Mysliveček [aus Ascanio in Alba von Mozart]?”). Responses were given in a yes/no format.

Musical expertise

We generated a composite indicator of musical expertise by taking the mean of four indicators: Gold-MSI scales Perceptual Abilities and Musical Training, and two items assessing how often participants attended live music events and how often they listened to music in the past (Mullensiefen, Gingras, Musil, & Stewart, 2014). The mean score on this composite indicator in our sample was M = 4.65 (SD = 0.89), and its correlation with age was small (r = .14, p = .071).

Procedure

Experimental sessions were conducted individually in a computer lab comprising 12 separate workstations and took about 1 hour for completion (including another task not reported here). Each participant wore closed-back studio headphones (Beyerdynamic DT-770 Pro, 80 Ω) and was seated in front of a computer screen at an individually comfortable distance separated from the other participants by sidewalls. Workstations were running Presentation^® (Neurobehavioral Systems Inc., Albany, CA) for controlling the presentation of textual and musical stimuli.

Before the start of the experiment, participants were randomly assigned to one of the four different experimental conditions. Participants first completed the Gold-MSI and questions on age, sex, and education. They then read an introduction describing the purported background of the study and were asked to listen carefully to the music to be presented (cover story and full text of the introduction can be found in the online supplemental material). They then read the manipulated description of composer and musical characteristics, and then listened to the musical piece via headphones. Immediately after listening to the music, participants completed a computerized questionnaire presenting the items on liking and perceived characteristics of the music heard, and questions on level of familiarity with composer and music. Finally, they were thanked and gave their written consent after dehoaxing.

Statistical analysis

For each outcome variable (liking and four factor scores of perceived musical characteristics), we implemented a 2 (composer) × 2 (writing mode) between-subjects full factorial ANOVA including age as a continuous covariate. Non-significant interactions were removed and models re-estimated to yield more parsimonious final models. If age showed any significant effects in a model, we additionally included musical expertise and any relevant interaction into that model to check if age effects were due to musical expertise. Age and musical expertise were mean-centered, so that main effects of composer and writing mode represent effects at average values of covariates. For illustration of significant effects, predictive margins were calculated and plotted from model predictions, with observations shown in underlying scatterplots or as means of binned variable intervals.²

All analyses were performed using Stata v14.2 (StataCorp LLC, 2017).

Results

The mean liking rating over all subjects and conditions was M = 4.6 (SD = 1.2, Min = 1, Max = 6). Sixty-one percent of participants chose one of the two highest liking categories, indicating that overall liking of the music was high.

Ten out of 86 (12%) participants who had to read texts with attributions to Mysliveček had listened to music by this composer before, and 2 out of 86 (2%) participants stated that they knew this particular piece of music already. In contrast, 83 out of 84 (99%) participants who were made believe that they had listened to a piece of music by Mozart declared that they have heard compositions by Mozart before. A small group of 14 out of 84 (17%) from this participant group declared that they were familiar with this specific piece of music.

Perceived musical characteristics

Results of the analyses of composer, writing mode, and age effect are shown in Table 1. Analyses revealed no significant main effects of writing mode and composer, or their interaction, on scale scores of QoAE, Vitality, or Emotional Expressivity. There were significant but rather subtle effects of age on perceived Affective Character and Vitality, with older participants perceiving higher levels of affective character (predicted margin at age 20 years = 4.55; at age 70 years = 4.98) and lower levels of vitality in the music heard (predicted margin at age 20 years = 5.87; at age 70 years = 5.47).

Table 1.

Results of ANOVAs with predictors composer, writing mode, and age, including significant interactions, on individual ratings of liking and perceived musical characteristics. The two-way interactions Composer × Writing mode and Writing mode × Age, and the three-way interaction were not significant in any model and are therefore omitted from all re-estimated models.

Source	Outcome variable
	Liking	Quality of Artistic Expressivity	Affective Character	Vitality	Emotional Expressivity
Composer	F(1, 165) = 0.64 p = .42	F(1, 165) = 1.8 p = .19	F(1, 166) = 0.1 p = .81	F(1, 166) = 0.6 p = .43	F(1, 166) = 0.2 p = .67
Writing mode	F(1, 165) = 10.6 p = .001 η² = .06	F(1, 165) = 0.0 p = .85	F(1, 166) = 12.8 p < .001 η² = .07	F(1, 166) = 0.7 p = .41	F(1, 166) = 2.6 p = .11
Age	F(1, 165) = 0.00 p = .97	F(1, 165) = 0.8 p = .36	F(1, 166) = 7.9 p = .006 η² = .05	F(1, 166) = 6.3 p = .013 η² = .04	F(1, 166) = 0.8 p = .37
Composer × Age	F(1, 165) = 7.3 p = .008 η² = .04	F(1, 165) = 5.8 p = .017 η² = .03
Variance explained	R² = .11	R² = .05	R² = .11	R² = .04	R² = .02

Writing mode showed a highly significant main effect on perceived Affective Character, with higher ratings when participants had read the expressive description before listening to the piece of music (Figure 1A). However, there was no significant main effect for composer, and no interaction between writing mode and composer, suggesting that writing mode affects perceived Affective Character of this music independently of composer attributions.

Figure 1.

Effects of writing mode, i.e., reading analytic or expressive texts prior to listening to music. 1(A): Predictive margins of responses on Affective Character scale scores with underlying scatterplot of observations. 1(B): Predictive margins of responses on liking ratings with underlying scatterplot of observations. Error bars indicate 95% confidence intervals.

Interestingly, results revealed a significant Composer × Age interaction for QoAE scores. As can be seen from Figure 2, perceived QoAE of the music heard differed significantly as a function of age and composer: While older participants did not show differences in how they perceived music attributed to Mozart or Mysliveček, younger adults between 20 and 40 years of age rated QoAE higher when the music was attributed to Mozart.

Figure 2.

Differences in scores on the scale QoAE at fixed values of age after reading texts on Mysliveček or Mozart prior to listening to music. (A) Predictive margins at a series of fixed values of age ranging from 20 to 70 at a step of 5 years, with underlying scatterplot of observations, both separately plotted for the two texts. Differences up to and including age 40 years were statistically significant at p < .05. (B) Contrasts of predictive margins showing differences between text conditions at fixed values of age with 95% confidence intervals. (C) Observed mean QoAE scale scores for five bins of age, with error bars indicating standard errors of the mean.

To check if the Composer × Age interaction was due to age differences in musical expertise, we included the composite indicator of musical expertise and its interaction with composer in the model. Results revealed a significant Composer × Expertise interaction, F(1, 163) = 5.0, p = .027, η² = .03, while there were no differences in QoAE scores at average levels of expertise F(1, 163) = 0.2, p = .68. Figure 3 shows that, similar to age, perceived QoAE was rated higher when the music was attributed to Mozart only if participants had a below-average level of musical expertise. Interestingly, this significant interaction was independent of the Composer × Age interaction, which still showed a significant effect in this model F(1, 163) = 4.5, p = .036, η² = .03.

Figure 3.

Differences in scores on the scale QoAE at fixed values of musical expertise after reading texts on Mysliveček or Mozart prior to listening to music. (A) Predictive margins at a series of fixed values of musical expertise ranging from 2.5 to 6.5 at a step of 0.5 scale score points, with underlying scatterplot of observations, both separately plotted for the two texts. Differences up to and including musical expertise scores of 4 were statistically significant at p < .05. (B) Contrasts of predictive margins showing differences between text conditions at fixed values of musical expertise with 95% confidence intervals. (C) Observed mean QoAE scale scores for five bins of musical expertise (low = 2.01–3.98, medium–low = 3.99–4.41, average = 4.42–4.91, medium–high = 4.92–5.46, high = 5.46–6.54), with error bars indicating standard errors of the mean.

Liking

We found no main effects of composer or age on how much participants liked the music they had heard. However, we found a significant main effect of writing mode as well as a significant interaction Composer × Age. Participants who had read an expressive description prior to listening liked the music more than participants who read an analytic description. The difference is illustrated in Figure 1B.

Similar to perceived QoAE, liking ratings differed significantly as a function of Composer × Age, as illustrated in Figure 4. Young adults between 20 and 35 years of age reported higher liking of the music when it was attributed to Mozart, whereas older participants between 40 and 70 years displayed no clear difference between the two conditions.

Figure 4.

Differences in liking ratings at fixed values of age after reading texts on Mysliveček or Mozart prior to listening to music. (A) Predictive margins at a series of fixed values of age ranging from 20 to 70 at a step of 5 years, with underlying scatterplot of observations, both separately plotted for the two texts. Differences up to and including age 35 years were statistically significant at p < .05. (B) Contrasts of predictive margins showing differences between text conditions at fixed values of age with 95% confidence intervals. (C) Observed mean liking ratings for five bins of age, with error bars indicating standard errors of the mean.

We included musical expertise and its interaction with age in the model to test if the Composer × Age interaction was due to expertise. However, neither musical expertise, F(1, 163) = 0.6, p = .44, nor its interaction with age, F(1, 163) = 0.7, p = .39, were significantly associated with liking ratings, while the Composer × Age interaction, F(1, 163) = 6.1, p = .014, η² = .04, was virtually unaffected by adjusting the model for expertise.

Discussion

Regarding our first hypothesis, results demonstrated significant differences in liking scores and in scores on the dimension Affective Character, dependent on the two different writing modes (analytic, expressive). Liking ratings were significantly higher if listeners had read the text written in the expressive writing mode when compared with the analytic writing mode (see Figure 1B). Furthermore, the mean Affective Character scale score difference between writing modes (see Figure 1A) indicates that participants who read the expressive text prior to listening rated the music as being more light, relaxed, soft, cheerful, and lyrical when compared with analytic writing mode.

There are several possible explanations for this result. The affective, imaginative, and florid language combined with the anthropomorphic character of the expressive writing mode might have evoked higher emotional engagement and empathy, which subsequently caused higher liking ratings compared with the more objective and formal language of the analytic writing mode. As noted by Thompson (2006, 2007), emotional engagement is a reliable predictor of overall enjoyment and highly related to listeners’ (positive) affective responses to music. The illustrative and narrative character of the expressive writing mode may have had a more supportive effect on the interpretive activity of the listeners than the analytic writing mode, facilitating higher engagement. This is supported by research showing that interpretive activity enhances recipients’ relationship with an art work on affective and cognitive dimensions (Cupchik, Shereck, & Spiegel, 1994).

Another explanation might be that the expressive writing mode caused higher liking ratings due to the induction of higher processing fluency, which is assumed to be associated with more positive evaluations (see Reber, Schwarz, & Winkielman, 2004). More specifically, high(er) processing fluency may have been evoked by the figurative language of the expressive writing mode, which provided an intuitively accessible content of images, exalting the imagination of the listeners through phrases. In contrast, the analytic writing mode was written in a more abstract style, using a sophisticated language including technical vocabulary from music theory that was presumably more difficult to understand. This may have caused the participants to concentrate more on single musical aspects during listening (e.g., musical form and texture) and thereby resulted in lower processing ease and fluency for this group of participants.

A more general supporting explanation for increased liking and altered perceived affective character of the music might be that the expressive writing mode caused more positive mood states as compared with the analytic writing mode. The induced positive mood may in turn have had an influence on the evaluation of the music in a positive direction (see Isen, Shalker, Clark, & Karp, 1978).

The effects of writing modes on liking and perceived affective character notwithstanding, musical characteristics in the dimensions QoAE, Vitality, and Emotional Expressivity have hardly been affected by the use of divergent writing modes. Hence, our results provide only partial support for the hypothesis that verbal notions about music can be treated as an interfering with the listener’s appreciation of musical characteristics (Wellmer, 2009). Although it is not clear why these other dimensions of musical characteristics were not affected, the cover story (historically informed performance practice) may have led the listener to focus more on the performance (quality) of the orchestra and the sound of its instruments rather than on the specific characteristics of the composition itself.

Perhaps the most striking outcome of the present study was that prestige effects were moderated by age, that is, authorship attributions to distinct composers (Mysliveček, Mozart) led to different liking ratings in younger participants only (see Figure 4). Our second hypothesis was only supported within the group of younger adults. Therefore, North and Hargreaves’ (2008) assumption that the attribution to Mozart would have an effect of prestige on the evaluative responses could only be confirmed for a subgroup of listeners for this specific piece. We found a very similar effect for the dimension of perceived QoAE (see Figure 2). This might be due to its proximity to Liking. Both measures share a highly evaluative character, indicated by items such as Interesting–Boring, Unimaginative–Imaginative, Superficial–Profound, Pleasant–Unpleasant, etc. (see full list of items in the online supplemental material) and by the correlation of Liking with QoAE scores (r = .69) that was clearly higher than with any of the other dimensions measured (.38 < r < .44).

While these results match those observed in earlier studies on prestige effects for the group of young adults (20–35 years), our study revealed clearly contradicting results with regard to the group of older participants. These older participants apparently had a high degree of field independence and were less susceptible to prestige suggestion than their younger counterparts. They may have benefitted from more experience in general as well as a long-term exposure to (classical) music and listening experience as regular concertgoers, as suggested by Crozier and Chapman (1981). Interestingly, these findings mirror those of a previous study by Bennett and Ginsborg (2018) that has examined audience reactions to program notes (although without a focus on prestige effects). The authors reported that “[m]ore experienced listeners were far more likely to reject the program note information in favour of their own interpretation” (Bennett & Ginsborg, 2018, p. 588).

Associated to that, ratings of older participants who supposedly acquired a considerable amount of knowledge during their lifespan about the classical repertoire, including the oeuvre of Mozart, may have been less high since the treatment text about Mozart contains the information that the opera Ascanio in Alba was written in 1771 when Mozart was (only) 15 years old (see full text in the online supplemental material). This explanation would also account for the younger participants being strongly biased by simply reading the name Mozart without further pondering on the composer’s sophistication. They probably engaged an “automatic pilot” of aesthetic evaluation, which would be in line with the classical definition of prestige effects in terms of: “If it’s Mozart, it must be good!”³

In addition, against the background of age-dependent effects of prestige, musical expertise suggests itself as having played an important role with regard to susceptibility: for example, listeners with a high level of musical expertise may have been less susceptible to prestige suggestion compared with listeners with a low expertise as proposed by Mittler (1976), and musical expertise may be expected to increase with age.

In the present study, however, neither the composite indicator of musical expertise nor its interaction with age were significantly associated with liking ratings for the different conditions. Hence, musical expertise, at least in the way it was measured in our study, cannot explain effects of prestige with regard to liking of the music. This appears to be counter-intuitive because we also observed that prestige effects were age-dependent, and it may be assumed that age is related to musical experience and knowledge of the classical repertoire.

Liking was measured by a single item in order to gather information about the individual’s appreciation of the musical piece as a whole. We assume that the responses to this item were primarily affective in character and strongly associated to participant’s individual taste. Therefore, one reason why level of expertise did not yield differences with regard to prestige effects might be that the scales of the composite indicator of musical expertise (taken from the Gold-MSI) measured musical expertise independently from musical styles and genres. Given that musical taste presumably varied across participants, we suppose that the specific expertise on classical music also varied as a function of musical taste and familiarity.

More precisely, we have to assume that older listeners were much more familiar with classical music than the younger adults from our sample, since musical taste (at least in the western world over the last decades) is highly age-specific and preferences for classical music styles increase with age (e.g., Bonneville-Roussy, Rentfrow, Xu, & Potter, 2013; Bonneville-Roussy, Stillwell, Kosinski, & Rust, 2017; Harrison & Ryan, 2010). These age trends have also been observed by surveys in the domain of consumer research. For example, sales figures from German sociodemographic music market research (GfK Consumer Panels) for 2016 showed that 92% of the volume of sales for classical music styles was caused by older consumers 40 years of age and above. In contrast, the group of younger adults from 20–39 years of age only caused 6% of the sales volume (see Bundesverband Musikindustrie, 2016, p. 31).

In contrast to the results on liking mentioned above, further analyses revealed a significant Composer × Expertise interaction similar to age for QoAE scale scores (see Figure 3). The perceived QoAE was rated higher when the music was attributed to Mozart in case participants had a below-average level of musical expertise compared with the condition when the music was attributed to Mysliveček. Although QoAE features a number of items with evaluative character, the semantic differentials are object-oriented in the sense that the items were related to certain musical characteristics of the musical piece.

Over the last decades, theories on decision making, reasoning, and social cognition have generated a number of dual-process theories of higher cognition (see Evans, 2008, 2009 for a review). The core version of it applies well to the two different tasks in question. The first type of process is generally described as being more or less automatic, fast in processing speed, low in effort, and high in processing capacity, while the second type of process is slow, controlled, high in effort, and limited with regard to processing capacity (Evans, 2009). The simple liking ratings used in our study correspond to the first processing type, as these affective responses were presumably made without effort on the basis of minimal cognitive engagement, since participants rather immediately knew how much they liked the music heard (see Zajonc, 1980). In addition, as mentioned earlier, liking ratings are also largely based on the individual’s musical taste and it is therefore harder to manipulate this kind of affective judgment by prior information.

In contrast to the fast and effortless type 1 process, the semantic differentials used in our study clearly correspond to the type 2 process. Responding to these semantic differentials with reference to the piece of music heard is a rather complex object description task involving more controlled processes and high effort, as participants were forced to concentrate on a variety of musical details and had to think about their decisions (e.g., weighing whether the music was more superficial or profound, unimaginative or imaginative, colorful or pale, etc.). As mentioned earlier, most of the items within the dimension QoAE were of highly evaluative character. However, in contrast to the liking ratings, these judgments were presumably less influenced by or based on individual musical taste. Moreover, participants who scored low on the scales of the composite indicator of musical expertise (e.g., perceptual abilities) may have been less confident with regard to their decisions similar to young(er) participants, meaning that their decisions were less stable. The assessment might have been more difficult for participants with lower music-related perceptual abilities and less musical expertise as compared with participants with more expertise.

In summary, these results agree with the findings of other studies, which showed that individual differences may play an important role with regard to susceptibility (e.g., Anglada-Tort & Müllensiefen, 2017; Chapman & Williams, 1976). Furthermore, our findings confirm previous findings with regard to the prestige effect, and they provide additional evidence suggesting that effects of prestige suggestion depend on age and experience.

Limitations and outlook

As a consequence of the highly controlled experimental setup, the current investigation has limited ecological validity due to the laboratory situation. For example, all participants were requested to read the text prior to music listening. In real life or, for example, in a real concert, not all listeners will read the program note prior to the start of the concert. Moreover, program notes or booklet texts are always part of larger social frames (Goffman, 1974). Future research should therefore focus on the investigation of framing effects, including both more ecologically valid listening scenarios (e.g., real concert settings) and a variety of pieces of music from different musical styles and genres.

Supplemental Material

POM812216_Online_Supplementary_Material – Supplemental material for If it’s Mozart, it must be good? The influence of textual information and age on musical appreciation

Supplemental material, POM812216_Online_Supplementary_Material for If it’s Mozart, it must be good? The influence of textual information and age on musical appreciation by Timo Fischinger, Michaela Kaufmann and Wolff Schlotz in Psychology of Music

Footnotes

Acknowledgements

The authors are grateful to Muralikrishnan, Claudia Lehr, Elena Felker, and Sophia Lehner for their help with data collection. We also want to thank Felix Bernoully for designing the flyers for participant recruitment.

Funding

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

Supplemental material

Supplemental material for this article is available online.

Notes

References

Anglada-Tort

Müllensiefen

(2017). The repeated recording illusion. Music Perception, 35(1), 94–117. doi:10.1525/mp.2017.35.1.94

Bashford

(2003). Not just ‘G’: Towards a history of the programme notes. In Musgrave

(Ed.), George Grove, music and Victorian culture (pp. 115–142). London: Palgrave.

Bennett

Ginsborg

(2018). Audience reactions to the program notes of unfamiliar music. Psychology of Music 46(4), 588–605. doi.org/10.1177/0305735617721339

Blohm

Menninghaus

Schlesewsky

(2017). Sentence-level effects of literary genre: Behavioral and electrophysiological evidence. Frontiers in Psychology, 8(1887). doi: 10.3389/fpsyg.2017.01887

Bonneville-Roussy

Rentfrow

P. J.

M. K.

Potter

(2013). Music through the ages: Trends in musical engagement and preferences from adolescence through middle adulthood. Journal of Personality and Social Psychology, 105(4), 703–717. doi:10.1037/a0033770

Bonneville-Roussy

Stillwell

Kosinski

Rust

(2017). Age trends in musical preferences in adulthood: 1. Conceptualization and empirical investigation. Musicae Scientiae, 21(4), 369–389. doi:10.1177/1029864917691571

Bundesverband Musikindustrie (2016). Jahresbericht 2016: Musikindustrie in Zahlen 2016. Berlin: Eigenverlag.

Chapman

A. J.

Williams

A. R.

(1976). Prestige effects and aesthetic experiences: Adolescents’ reactions to music. British Journal of Social and Clinical Psychology, 15, 61–72.

Crozier

W. R.

Chapman

(1981). Aesthetic preferences: Prestige and social class. In O’Hare

D. P. A.

(Ed.), Psychology and the arts (pp. 242–278). New Jersey: Humanities Press.

10.

Cupchik

G. C.

Shereck

Spiegel

(1994). The effects of textual information on artistic communication. Visual Arts Research, 20, 62–78.

11.

Entman

R. M.

(1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43, 51–58.

12.

Entman

R. M.

(2003). Cascading activation: Contesting the White House’s frame after 9/11. Political Communication, 20(4), 415–432.

13.

Evans

J. S.

(2008). Dual-processing accounts of reasoning, judgment, and social cognition. Annual Review of Psychology, 59, 255–278. doi:10.1146/annurev.psych.59.103006.093629

14.

Evans

J. S. B. T

. (2009). How many dual process theories do we need: One, two or many? In Evans

Frankish

(Eds.), In two minds: Dual processes and beyond (pp. 33–88). New York, NY: Oxford University Press.

15.

Goffman

(1974). Frame analysis: An essay on the organization of experience. Boston, MA: Northeastern University Press.

16.

Hargreaves

D. J.

(1986). The developmental psychology of music. Cambridge, UK: Cambridge University Press.

17.

Harrison

Ryan

(2010). Musical taste and ageing. Ageing and Society, 30(04), 649–669. doi:10.1017/s0144686x09990778

18.

Isen

A. M.

Shalker

Clark

Karp

(1978). Affect, accessibility of material in memory, and behavior: A cognitive loop? Journal of Personality and Social Psychology, 36, 1–12.

19.

Kahneman

Tversky

(1979). Prospect theory: An analysis of decision under risk. Econometrica, 47, 263–291.

20.

Kahneman

Tversky

(1984). Choices, values, and frames. American Psychologist, 39(4), 341–350.

21.

Kirk

Skov

Hulme

Christensen

M. S.

Zeki

(2009). Modulation of aesthetic value by semantic context: An fMRI study. NeuroImage, 44(3), 1125–1132. doi:10.1016/j.neuroimage.2008.10.009

22.

Kroger

Margulis

E. H.

(2016). “But they told me it was professional”: Extrinsic factors in the evaluation of musical performance. Psychology of Music, 45(1), 49–64. doi:10.1177/0305735616642543

23.

Lauring

J. O.

Pelowski

Forster

Gondan

Ptito

Kupers

(2016). Well, if they like it … Effects of social groups’ ratings and price information on the appreciation of art. Psychology of Aesthetics, Creativity, and the Arts, 10(3), 344–359. doi:10.1037/aca0000063

24.

Levin

I. P.

Schneider

S. L.

Gaeth

G. J.

(1998). All frames are not created equal: A typology and critical analysis of framing effects. Organizational Behavior and Human Decision Processes, 76(2), 149–188.

25.

Lin

Yao

(2017). The impact of accompanying text on visual processing and hedonic evaluation of art. Empirical Studies of the Arts, 36(2), 180–198. doi:10.1177/0276237417719637

26.

Margulis

E. H.

(2010). When program notes don’t help: Music descriptions and enjoyment. Psychology of Music, 38(3), 285–302. doi:10.1177/0305735609351921

27.

Margulis

E. H.

Kisida

Greene

J. P.

(2015). A knowing ear: The effect of explicit information on children’s experience of a musical performance. Psychology of Music, 43(4), 596–605. doi:10.1177/0305735613510343

28.

Mittler

G. A.

(1976). An instructional strategy designed to overcome the adverse effects of established student attitudes toward works of art. Studies in Art Education, 17(3), 13–31.

29.

Mullensiefen

Gingras

Musil

Stewart

(2014). The musicality of non-musicians: An index for assessing musical sophistication in the general population. PLoS ONE, 9(2), e89642. doi:10.1371/journal.pone.0089642

30.

North

A. C.

Hargreaves

D. J.

(2005). Brief report: Labelling effects on the perceived deleterious consequences of pop music listening. Journal of Adolescence, 28(3), 433–440. doi:10.1016/j.adolescence.2004.09.003

31.

North

A. C.

Hargreaves

D. J.

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

32.

Pan

Kosicki

G. M.

(1993). Framing analysis: An approach to news discourse. Political Communication, 10(1), 55–75. doi:10.1080/10584609.1993.9962963

33.

Reber

Schwarz

Winkielman

(2004). Processing fluency and aesthetic pleasure: Is beauty in the perceiver’s processing experience? Personality and Social Psychology Review, 8(4), 364–382. doi:doi:10.1207/s15327957pspr0804_3

34.

Schaal

N. K.

Bauer

A.-K. R.

Müllensiefen

(2014). Der Gold-MSI: Replikation und Validierung eines Fragebogeninstrumentes zur Messung Musikalischer Erfahrenheit anhand einer deutschen Stichprobe. Musicae Scientiae, 18(4), 423–447. doi:doi:10.1177/1029864914541851

35.

Spitzer

(1987). Musical attribution and critical judgment: The rise and fall of the Sinfonia Concertante for Winds, K. 297b. The Journal of Musicology, 5(3), 319–356. doi:10.2307/763696

36.

Tewinkel

(2016). Muss ich das Programmheft lesen? Zur populärwissenschaftlichen Darstellung von Musik seit 1945. Kassel: Bärenreiter.

37.

Thompson

(2006). Audience responses to a live orchestral concert. Musicae Scientiae, 10(2), 215–244. doi:10.1177/102986490601000204

38.

Thompson

(2007). Determinants of listeners’ enjoyment of a performance. Psychology of Music, 35(1), 20–36. doi:10.1177/0305735607068886

39.

Thorau

(2013). Werk, Wissen und touristisches Hören. Popularisierende Kanonbildung in Programmheften und Konzertführern. In Pietschmann

Wald-Fuhrmann

(Eds.), Der Kanon in der Musik – Theorie und Geschichte. Ein Handbuch (pp. 535–561). München: edition text+kritik.

40.

Thorau

. (in press). “What ought to be heard”. Touristic listening and the guided ear. In Thorau

Ziemer

(Eds.), The Oxford handbook of music listening in the 19th and 20th centuries (pp. 207–227). New York: Oxford University Press.

41.

Tousignant

Bodner

G. E.

(2014). Context effects on beauty ratings of photos: Building contrast effects that erode but cannot be knocked down. Psychology of Aesthetics, Creativity, and the Arts, 8(1), 81–86. doi:10.1037/a0034942

42.

Umanath

(2016). Age differences in suggestibility to contradictions of demonstrated knowledge: The influence of prior knowledge. Neuropsychology, Development, and Cognition. Section B, Aging, Neuropsychology and Cognition, 23(6), 744–767. doi:10.1080/13825585.2016.1167161

43.

van Paasschen

Bacci

Melcher

D. P

. (2015). The influence of art expertise and training on emotion and preference ratings for representational and abstract artworks. PLoS ONE, 10(8), e0134241. doi:10.1371/journal.pone.0134241

44.

Vuoskoski

J. K.

Eerola

(2015). Extramusical information contributes to emotions induced by music. Psychology of Music, 43(2), 262–274. doi:10.1177/0305735613502373

45.

Wagner

Klein

Hanich

Shah

Menninghaus

Jacobsen

(2015). Anger framed: A field study on emotion, pleasure, and art. Psychology of Aesthetics, Creativity, and the Arts, No-Pagination Specified. doi:10.1037/aca0000029

46.

Waxman

(2012). Prefacing music in the concert hall: Program books, composer commentaries, and the conflict over musical meaning (Unpublished doctoral dissertation). New York University, New York.

47.

Wellmer

(2009). Versuch über Musik und Sprache. München: Hanser.

48.

Zajonc

R. B.

(1980). Feeling and thinking. Preferences need no inferences. American Psychologist, 35(2), 151–175.

49.

Ziv

Moran

(2006). Human versus computer: The effect of a statement concerning a musical performance’s source on the evaluation of its quality and expressivity. Empirical Studies of the Arts, 24(2), 177–191. doi:10.2190/E4EN-1X32-KUU1-LDHT

Supplementary Material

Please find the following supplemental material available below.

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

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

0.00 MB

0.46 MB