Der Dichter spricht. PDF scanned by nlib. Kinderszenen, Op. Creative Commons Attribution-NonCommercial 4. Creative Commons Attribution 4. Creative Commons Attribution 3.
|Published (Last):||7 April 2016|
|PDF File Size:||3.17 Mb|
|ePub File Size:||12.38 Mb|
|Price:||Free* [*Free Regsitration Required]|
Performance Science View all 4 Articles. We explore the concept that artists perform best in front of an audience. According to the evaluation by listeners, the recordings performed in front of an audience sounded better, suggesting that the presence of an audience enhanced or facilitated the performance.
In Study 2, we analyzed pianists' durational and dynamic expressions. In Study 3, we explored a model that explained listeners' responses induced by pianists' acoustical expressions, using path analyses.
The final model indicated that the cross-sectional variation of the duration and that of the dynamics determined listeners' evaluations of the quality and the emotionally moving experience, respectively. Music has been played to an audience for millennia. Studio recording of music is a relatively new phenomena. Listeners who prefer live to studio recordings also believe that they can feel performers' passions from the live recordings Badal, Do recordings made in front of an audience actually sound better for listeners than those made alone?
If yes, how do these recordings differ acoustically, and how exactly do such differences determine listeners' experiences of the two types of recording? We explored these ideas in the present study. Researchers in performance science have explored factors determining performers' expressions. For example, Hargreaves et al. Previous studies have empirically confirmed that styles Baroque, Romantic, Modern, e. In a competition Yoshie et al.
These psychological and physiological stresses degrade the artistic quality of the performance e. This suggests that the quality of music performance can also be enhanced while performed live, at least, when skilled performers play a familiar piece. The purpose of the present study was to test this hypothesis in piano performance. The present research consisted of three studies. In Study 1, we tested whether social facilitation would exist in piano performance by means of listeners' evaluations between the audience-present and the audience-absent recordings made by 13 pianists.
Studies 2 and 3 were conducted to identify pianists' acoustical expressions differing between the two recordings and to determine how acoustical differences would explain listeners' evaluations, respectively. As performance practice for a piece from the Romantic period Palmer and Halford, , pianists employ the overall tempo rubato i.
We explored the mechanism of social facilitation in pianists' performances of this particular piece. Figure 1. Section IDs boxed letters are added for this Figure. This image was reproduced by the first author. A quality scale measures the overall quality of performance including both technical and expressive aspects. When this scale is used to assess multiple versions of the same piece, listeners' evaluations may reflect their preferences Clarke, Music psychologists consider this experience different from the listener's perception of the quality of the performance e.
We hypothesized that music recorded in front of an audience would sound better and emotionally move listeners more than the audience-absent recordings, in line with the theory of social facilitation Zajonc, ; Bond and Titus, ; Strauss, Participants were not majored in music but had at least 9 years of classroom music instruction including music appreciation as a compulsory school education.
Participants were categorized into three groups based on the years of musical training: least experienced i. Participants received JPY approximately U. The informed consent was obtained from each participant prior to the experiment, following ethical standards suggested by American Psychological Association They started to play the piano between ages 4 and 6.
We asked pianists to practice these pieces at least for 1 month before the recordings. Recordings took place in a small auditorium with the maximum capacity of , equipped with a grand piano GP, Boston.
The sound pressure level of background noise was The piano was tuned professionally within 1 week before each pianist's recording. On the day of recording, pianists rehearsed each piece as many times as they wished before the performance. Subsequently, each pianist performed six pieces in a random order specified by the experimenter H.
Performers could re-record the performance as many times as they wished in the audience-absent context. The performances were stereo-recorded onto a multi-track recorder R24, Zoom using a microphone NT4, Rode. The performance portion of the recording lasted approximately 45 audience-present and 30 min audience-absent.
The first author confirmed by listening to all the recordings that no noise from the audience was audio-recorded. Ten weeks after they attended the live recording of the pianist's performances, participants returned to the same auditorium, and listened to both the audience-present and the audience-absent recordings of six pieces played by the same pianist, individually or in a group of 2— We considered the insertion of 10 weeks after the participant's initial exposure to the live performance to be enough to eliminate a possibility of a mere-exposure effect i.
The order of recording contexts was counterbalanced. The order of stimuli in the first block was randomized for each pianist and the same stimulus order was used in the second block. In other words, each participant listened to both the audience-present and the audience-absent recordings in the same order. After evaluating all the pieces, participants provided their background information i. Table 1 shows the mean scores of the quality and the emotionally moving scales based on training.
We conducted a 3 i. Table 1. The mean scores of quality A and emotionally moving experience B by the least, the moderately, and the most experienced listeners. The above results confirmed that listeners evaluated the audience-present better than the audience-absent recording regardless of their level of musical training, as found in their quality ratings. As Yoshie et al. Thus, the general theory of audience effect Bond and Titus, ; Strauss, seems applicable to music performance: The difficulty of the task determines whether the presence of the audience leads to social facilitation or inhibition.
It is possible, then, that the mechanism of social facilitation in the emotionally moving experiences is different from that in the perception of quality of the performance.
We explore this issue by incorporating the acoustical measurements of the performances in Studies 2 and 3. One of the goals of music performance is communication of the structure of the piece, reflecting the composer's idea Clarke, ; Palmer, ; Friberg and Battel, Performers can communicate with the audiences by highlighting their expressions along with the musical structure. The degree of lengthening is the greatest at the end of the piece i.
Despite the valuable contributions made by Repp , , his original analyses treated each note as a different variable, ignoring that notes being closer in the score are more statistically related than those being farther away Almansa and Delicado, To solve this problem, Almansa and Delicado applied their proposed time-series analyses to the expressive timing data of Repp's , measuring the duration for each eighth note of the piece while treating each performance as a continuous function of one variable.
Using functional principal components analysis Ramsay et al. The first component The second component The remaining three components explained more local features of the timing variation 5. In the present study, we examined whether the same two components could be replicated in addition to how pianists differentiated each component between the recording contexts.
We also conducted time-series analyses on pianists' dynamic expressions. If this were true, we would extract the same components for both the durational and the dynamic expressions, and pianists' differentiations between the recording contexts would also be consistent.
Participants i. We would describe measurement and computation of parameters below. For the present study, we measured duration and dynamics in the following. As for dynamics, the A-weighted sound level dBA per 3. Note that the sound level at the softest keystroke is not influenced by the decay of piano tone unlike the minimal sound level within a beat.
Both the minimal and the maximal keystrokes can be considered to be expressive manipulations controlled by the pianists. For this reason, we believe that the within-beat difference of these values i. The mean of the beat duration and that of the dynamic range represent the overall tempo and dynamic range, respectively.
We analyzed the variations of these parameters as well. Our measurement neither used MIDI format nor treated an eighth note as a unit of analysis, as did previous studies Repp, ; Almansa and Delicado, This was because a MIDI console was unavailable on our Boston piano, and we decided to go in line with Repp , who identified the onset of each beat manually, as mentioned above. Moreover, we believed that pianists would manipulate their expressions based not on the shortest note but on the beat of a piece i.
To confirm the reliability of the identification, two volunteers independently confirmed the accuracy of the first author's measurements of the beat duration for all the recordings. We computed mean, coefficient of variation, and cross-sectional variation of the duration and the dynamic range for each performance Table 2. To examine effects of the recording context on these parameters, we conducted permutation paired t -tests.
A permutation test is an alternative way to examine differences in population parameters in a non-parametric fashion, so that we do not need to make any assumption about the sampling distribution e. Moreover, permutation tests allow us to use raw data unlike conventional non-parametric methods e.
We computed a t -statistic of the obtained observations t obt , and then, the sampling distribution of the t -statistic was generated by means of 10, iterations of permutated data. A p -value was obtained by computing a proportion of the iterations that was equal to or greater than the actual grouping of the data. However, the calculation of the parameters from the beginning to the end of the piece might lose information such as pianists' particular expressions at structurally important locations e.
In addition, such time-series expressions might differ as a function of pianist. To solve these problems, we conducted a functional principal components analysis Ramsay et al. Table 2. The mean M , the coefficients of variation CV , and the cross-sectional variations CSV of the duration and the dynamic range for the audience-present and the audience-absent conditions across 13 pianists. The following analyses were based on the method by Almansa and Delicado First, we calculated the smoothed beat duration by computing the cumulative value from the beat duration data.
Kinderszenen, Op.15 (Schumann, Robert)
Schumann wrote 30 movements for this work but chose 13 for the final version. He told his wife Clara that the "thirty small, droll things", most of them less than a page in length, were inspired by her comment that he sometimes seemed "like a child". He described them in as "more cheerful, gentler, more melodic" than his earlier works. Movement No.
Home Help Search. The only recording in the world of Schumann's "Traumerei" at the correct tempo Read times. Member Posts: But this is not something that is written for us by the composer at all. Often she joked over Schumann that he seemed and behaved like a child. And there's nothing wrong with: many good people remain to be children until their last days - it had been noticed for long. Richard Cock.