Expert pianists’ practice perspectives: A production and listening study

Abstract

The purpose of this study was to investigate how professional pianists practice music for a concert, and whether their individual cognitive orientations in such practice processes can be identified accurately from the resulting performances. In Study I, four pianists, previously found to be skilled music memorizers, practiced and performed a short piece by André Jolivet over the course of two weeks, during which their practice strategies were studied using semi-structured interviews, and analyses of practice diaries, practice activities, and eye-movement data. The results indicate that the pianists used similar basic strategies but had different cognitive orientations, here called “practice perspectives,” consistent with each individual, in that they focused on different kinds of information while practicing. These practice perspectives may be related to skills and habits in using imagery and music analysis, as well as to professional and educational background. In Study II, 34 piano teachers listened to recordings of the concert performances and evaluated them against 12 statements representing the four practice perspectives identified in Study I. The results did not support the prediction that practice perspectives would be correctly detected by listeners. Nonetheless, practice perspectives can be used to highlight potentially vast differences between the ways in which individual professional classical musicians conceptualize music and make it meaningful to themselves and others. They could be used in the context of music education to increase musicians’ knowledge of different practice strategies and the ability to develop their own preferred working methods.

Keywords

Expertise individual differences music practice piano performance musical interpretation

In the tradition of Western classical music, musicians usually learn their repertoire from notated scores, but the same scores can be performed by different musicians in highly individual ways (e.g., Repp, 1992). Expression in performance may depend, for example, on the performer’s technical skills, structural interpretation of the music, personal style, mood, expressive intentions, and/or performer-audience interaction (Juslin, 2003). Like any human interaction, musical performance can also be thought of as conveying some of the communicator’s intentions and personal features. Hence, while musicians’ individual “fingerprints” may be recognized computationally even from inexpressive scale playing (Van Vugt et al., 2013), musical performances may also show consistencies due to more intentional shaping of expressive parameters such as timing (Repp, 1995), dynamics (Repp, 1996), and timbral nuances (Bernays & Traube, 2014). Such connections between the musicians’ intentions and performance parameters (e.g., Nakamura, 1987; Palmer, 1989) suggest that during practice, musicians build a detailed mental representation of the piece and how they want to perform it. Performances are thus controlled through imagery and anticipatory models (see Altenmüller & Shinichi, 2016; Keller, 2012). Despite some spontaneity in live situations, performers often use prepared performance cues—landmarks in the mental map of the piece that the musician attends to during performance (see Chaffin et al., 2016; Ginsborg et al., 2012)—to guide their performances. Overall, a performance might be viewed as a result of practice strategies—the thoughts and behaviors in musicians’ practice “intended to influence their motivational or affective state, or the way in which they select, organize, integrate, and rehearse new knowledge and skills” (Jørgensen, 2004, p. 85).

Musicians are known to differ qualitatively in their approaches to interpretation and practice (Hallam, 1995; Héroux, 2018; Hultberg, 2008). Notated scores offer the possibility for different perspectives to be taken on their meaning. Musicians can, for example, choose a reproductive approach, seeing the score as a normative document, or take a more exploratory approach, relying on their own judgments of expressive conventions (Hultberg, 2000); the musician’s choice of approach may depend on their personal values (see Héroux, 2018). One musician might prefer to focus on the structure of the music, while another will approach the same piece through emotion and symbolic thinking (Dos Santos & Hentschke, 2011; see also Héroux, 2016). However the musician prefers to approach the notated score, their practice perspective is likely to be manifest in their practice strategies—that is, in the musician’s thoughts and actions while preparing a piece for a performance. With this in mind, the first of our two research questions addressed the respects in which skilled classical musicians might vary in their approaches to learning and practicing the same piece of music for a concert. For instance, musicians’ approaches to practice might conceivably differ in terms of their attention to the analysis of musical structure and/or interpretative choices, their conscious use of memorizing strategies, and so on.

Different approaches to private instrumental practice could obviously vary in their effectiveness, but this does not mean that they necessarily leave traces that can be discerned by those who listen to the performance, although such a link between cause and effect is conceivable (Watson, 1995, p. 14). We know that musicians are capable of communicating basic emotions in ways that affect performance parameters and are identifiable by listeners (Gabrielsson & Juslin, 1996; Juslin & Laukka, 2003). Experienced listeners may even be able to detect different, moderately independent components of expression in a performance (Juslin et al., 2001). The extent to which such differences can be traced to musicians’ cognitive orientations during practice is, however, unclear. Our second research question was whether experienced listeners can detect, when listening to musicians’ performances, the practice strategies they used [and therefore, potentially, their cognitive orientations].

In order to investigate how musicians’ cognitive orientations might come across to experienced listeners in concert performance, we undertook two interlinked studies using an exploratory sequential mixed methods design, with successive qualitative and quantitative components (Creswell & Creswell, 2018). For Study I, we chose a multiple case study approach (Baxter & Jack, 2008; Yin, 2018). Four pianists prepared to play the same piece from memory in a concert. We gave them the task of memorizing and performing a piece of post-tonal music, assuming that this would encourage them to use creative learning strategies, and that the piece’s relative lack of identifiable emotional connotations might elicit a broader palette of expressive choices than would be the case with tonal music. To increase the construct validity of our analysis of practice strategies, we used multiple sources of evidence (Yin, 2018). Practice strategies were studied using qualitative analyses of interview and practice diary data gathered at different stages of the two-week practice period. We also complemented interview data related to initial silent reading and first instrumental practice with analysis of eye-movement videos and practice data, respectively. Previously, both types of data have been used successfully as indicators of musicians’ thought processes while practicing (see Chaffin et al., 2003; Ginsborg et al., 2012; Héroux & Fortier, 2014; Loimusalo et al., 2019). In Study II, 34 piano teachers listened to recordings of the live performances given at the concert by the four pianists, and evaluated the extent to which the unique and consistent features of each pianist’s practice—referred to in this article as practice perspectives—were reflected in the performances. We used principal component analysis and linear mixed-effect models to analyze the listeners’ reactions to the performances in the light of the pianists’ practice strategies, as identified in Study I.

Study I

Learning to perform a post-tonal piece from memory: A case study of four pianists

Method

Participants

We used purposive sampling of participants based on an earlier study by Loimusalo & Huovinen (2018). We selected the four best music memorizers from that study, based on their total recall rates on the grounds that these participants were the ones most likely to demonstrate effective and interesting learning strategies. We will refer to the participants using the pseudonyms Terry, Finley, Kim, and Morgan, and feminine pronouns. Their mean age was 40.3 years (SD = 14.8), and they were all experienced musicians who had studied for 7–10 years at institutions of higher musical education.

Musical material

We selected “La vache” by André Jolivet (1905–1974) from the suite Mana for solo piano (1935) (see Figure 1) because of its suitable post-tonal structure. In addition, it was unlikely to be known to participants (and indeed they were not familiar with it before taking part in the study), as works by Jolivet are performed in public so rarely in Finland. Although the piece is rhythmically complex in places, it is short and therefore considered capable of being performed from memory after two weeks of practice. We did not wish participants to search for information about the piece or listen to recordings of it, so we removed potential identifiers from the score by covering the beginning of the next movement of the suite and replacing the French and Russian titles of the piece with their translation into Finnish, “Lehmä” (“The cow”).

Figure 1.

“La vache” from Mana, a suite for solo piano by André Jolivet (1986), composed in 1935 published with permission given by © Editions Jobert, Paris.

Procedure

The four participants were asked to prepare the piece for performance in a small “concert event” over the course of two weeks. During this time, the first author met each participant three times. Session 1 took place at the start of the project on Day 1, Session 2 midway through the practice period on Day 8, and Session 3 included the concert on Day 15. In Session 1 each participant read the piece silently, after which a semi-structured one-to-one interview took place, and then practiced the piece on the piano, after which they were interviewed again. One-to-one semi-structured interviews were also conducted with all the participants in Session 2 following the first week’s practice. In Session 3, focus group interviews were conducted with two participants at a time after the concert, on the assumption that insights into their practice and artistic decisions would emerge from the opportunity for participants to talk with another expert pianist, familiar with the piece, as well as the first author. All three sessions were video- and audio-recorded. Participants’ eye movements were recorded in Session 1 during the initial silent reading (eye-tracking system Dikablis 2.0: field-camera resolution 768 x 576, 25 fps; eye-camera resolution 384 x 288, 25 fps) to validate the data gathered in the first interview. Institutional guidelines for ethical practice were followed throughout the research process. We subsequently analyzed data gathered in four stages of practice:

Silent reading (Session 1). In the beginning of the session, the participant gave her informed consent to take part in the study. After this, the researcher conducted 4-point calibration of the camera used for recording eye movements. Next, the researcher instructed the participant to “silently get acquainted with the following piece,” and revealed the score, allowing the participant to study it for two minutes in silence. The silent reading stage before playing and hearing the piece played aloud was added to the experiment on the assumption that this stage would reveal different ways of forming mental representations on the basis of reading the notation. During the silent reading, the participant’s eye movements were recorded. After silent reading, the researcher covered the score and began a short semi-structured interview (8–10 minutes), designed to examine the participant’s recall, with the prompt, “Tell me about the piece.” When the participant had done so, the researcher uncovered the score again, asking “What did you do while silently reading the score?”

First instrumental practice (Session 1). Next, the researcher invited the participant to acquaint herself with the piece for 10 minutes (nine minutes in the case of Finley, due to the researcher’s error) at a Yamaha C7 grand piano, before leaving the room so the participant could practice freely. On her return, the researcher conducted a second semi-structured interview with the same two prompts as in the first. Emerging themes were elaborated in both interviews through follow-up questions and answers. The total duration of the first session varied from 50 to 60 minutes.

Midway (Session 2). After a week in which the participants could practice as much or as little as they liked, the second session began with the participant performing the piece to the researcher from notation. Next, the researcher conducted a semi-structured interview asking the participant, first, to “tell me about the piece,” and then to describe “what was done during the first week in terms of practice.” Participants were encouraged to demonstrate at the piano. Again, emerging themes were elaborated through follow-up questions and answers.

Concert (Session 3). Two concert events were organized in which each participant performed the piece in the presence of the researcher, one other participant, and a small audience consisting of three or four other researchers. After both of the participants had performed, the audience left and the researcher conducted a focus group interview with the two participants in which they were asked to discuss the piece and to reflect upon their learning and memorizing strategies during the second week of the practice period, and on the whole process.

Participants also documented their practice activities during both practice weeks in diaries in which they were asked to note the date and time of practice, and to “describe as accurately as possible what you have done: how did you practice, what kinds of thoughts have you been thinking, what passages of music have you been processing/practicing, and how have you done it? Report also whether your practice happened with or without an instrument. If without, report whether it was intentional or whether you noticed processing the music more non-consciously.” Practice diaries complemented interviews in two ways. Practice strategies might not come to the participant’s mind while she was being interviewed, but they could be identified from diary entries. Furthermore, completing practice diaries could make participants more aware of their thoughts and behaviors, helping them to remember the strategies they had used.

Data analysis

Our strategy for analyzing the results of Study I was threefold. First, we wanted to identify and categorize participants’ practice strategies on the basis of their own spoken and written accounts (interview and diary data). Second, we wanted to validate their self-reports by comparing them with the pianists’ observed activities. Specifically, we compared video recordings of their eye movements while reading the score silently and audio-recordings of their first practice at the piano with the corresponding self-reflective accounts from the first session. Third, we hoped to condense the participants’ various strategies into descriptions of individual practice perspectives, asking how, and to what extent, they differed from each another. The first author analyzed all three types of data gathered in Study 1.

Practice strategies

First, interview data from all four stages were transcribed verbatim. The transcripts and practice diaries were read through so as to gain a first impression of their content, both spoken and written. In response to the two interview prompts, the participants described their “practice and reading activities,” and “the piece.” They also spoke about “the composer” and “the composer’s intentions and solutions,” their own “personal characteristics,” and “what [they] intended to do with the piece in future practice.” While the transcripts and practice diaries were being read for a second time, it was concluded that, irrespective of whether the music was being read silently or practiced on the piano, the participants’ reported activities were mostly cognitive in nature, involving processes such as thinking, reasoning, analyzing, reflecting, encoding, and imagining. It was also observed that the participants’ observations on or impressions of the piece often revealed the kinds of strategies they had used (e.g., while reading silently, observations of the hands crossing each other revealed motor imagery, descriptions of the music’s aural character indicated aural imagery). In data deriving from both the interviews and the practice diaries, comments referring to practice/reading, the piece, and the composer’s choices were interpreted as practice strategies, and labeled with descriptions of their core content. A total of 45 practice strategies was found, and grouped into eight “families”: temporal, analytical, aural, embodied, interpretational, narrative/associative, memory, and reading strategies (see Appendix 1). On this basis a coding scheme and instructions were developed. Finally, the interviews and practice diaries were reread, and participants’ statements were assigned to the main categories and subcategories of the coding scheme.

Silent reading processes

A more objective account of the participants’ first encounters with the score was obtained from analysis of their eye movements while reading silently. The video recordings were viewed several times using Dikablis 2.0 software. Vertical dislocations of participants’ fixations, slightly above or below the staves, seemed to occur occasionally, although such calibration errors were usually systematic within each recording. Owing to a calibration error, there was also some systematic data loss in Morgan’s video at the last line of the first page. The videos were not considered sufficiently accurate for a detailed analysis of eye movements to be made. Nevertheless, they provided rough but useful supplementary information concerning the general reading processes. Silent reading processes were coded by interpretatively allocating a number to each second of reading, each number referring to one of the 20 half-bars of the score indicated by bar lines or dashed bar lines—henceforth abbreviated as “bars” (see Figure 1).

Initial practice activities

The content of the participants’ audio-recorded activities while practicing on the piano in Session 1 was coded qualitatively, adapting a similar segmentation used, for example, in Chaffin et al. (2003). The method involved four steps. First, the recordings were inspected and each second was coded and entered on a spreadsheet in terms of its “location in the music” (i.e., bar number) or as “silence.” Second, the recordings were broken down into segments of activity. This involved identifying the parts of the recording during which the participant was continuously occupied by a particular kind of activity, described by codes such as “playing the piece through,” “repeating the pattern of the last 16th notes [bar 19] four times in different tempi,” “playing from the beginning of bar 16 to the rest in bar 18,” and “holding the keys down.” Third, similar activities were grouped under more general descriptions (e.g., “linear reading while playing,” “detailed reading,” “staying on a chord,” “break in the music”). Fourth, the coded practice activities were grouped into the following major categories and subcategories so that each segment of practice could be assigned to one of them:

Linear practice activities included two subcategories: holistic practice denoted playing the piece from beginning to end without interruptions; forward-oriented practice included playing forward without interruptions for at least seven beats (the length of one original bar).

Fragmented practice activities included playing one single short pattern of less than seven beats. If a pattern was played once, after which only a part of it was repeated, both patterns were coded as fragmented.

Repeated practice activities included repeating the same short pattern (from one to seven beats) more than once. The first occurrence of the pattern was coded as fragmented, and subsequent occurrences of the same pattern were coded as repeated.

Experimental practice activities designated practice in which a pattern was not performed as written, but the notes were played, for example, as cluster or an arpeggio, different notes were explored back and forth, or keys were depressed and held down, thereby interrupting the chosen practice tempo.

Silent practice activities involved a break in the playing not traceable to a rest written in the score.

Results: Four practice perspectives

For each participant, data gathered in each of the four practice stages was coded for the presence of the practice strategies identified in the material. The distribution of the 45 strategies across the four participants’ practice is shown in Appendix 2. Two general observations can be made. First, all the participants used most of the families of practice strategies—temporal, analytical, embodied, interpretational, and memory—while individual participants used aural and narrative strategies to different extents. Second, they used some strategies at particular stages of practice, such as overviewing the visual text in the first stage. Our first research question addressed the respects in which skilled classical musicians might vary in their approaches to learning and practicing the same piece of music for a concert, so our analysis focused on differences between individuals’ practice strategies. We therefore identified those families of practice strategies that were consistent (i.e., present at all stages) for each participant as well as characteristic of that participant in comparison with the others, interpretatively condensing them as the pianists’ practice perspectives, that is, their characteristic cognitive orientations while practicing. The four participants’ practice perspectives, summarized and labeled in Table 1, are presented in the following as narratives with descriptive examples from the data. Table 2 presents the percentage of each practice activity during each participant’s practice of the piece on the piano.

Table 1.

Summary of the four pianists’ practice perspectives and their characteristic strategies.

Pianist	Characteristic and consistent strategies	Family	Practice perspective
Terry	Local narrativization	Narrative/associative	Creative-narrative
	Global narrativization	Narrative/associative
	Emotional associations	Narrative/associative
	Visualizing actors or scenery	Narrative/associative
	Verbal associations	Narrative/associative
	Verbal mnemonics	Memory
	Visual mnemonics	Memory
Finley	Detailed study of rhythm	Temporal	Temporal-normative
	Focus on rhythmic/temporal character	Temporal
	Device-assisted systematic temporal work	Temporal
	Aural imagery of rhythm	Temporal
	Listening to the direction of pitch elements	Aural
	Solving practical problems	Embodied
	Segmented reading	Reading
Kim	Motivic analysis	Analytic	Aural-atmospheric
	Listening to pitch elements	Aural
	Listening to aural atmosphere	Aural
	Metaphoric expressions	Narrative/associative
Morgan	Motivic analysis	Analytic	Compositional-analytic
	Analysis of pitch structure	Analytic
	Thoughts on the composer’s intentions	Analytic
	Listening to compositional structure	Aural

Table 2.

Prevalence of practice activities during the first instrumental practice (% of time).

Pianist	Terry	Finley	Kim	Morgan
Linear practice	45.5	29.3	37.2	50.2
Fragmented practice	17.7	23.0	12.2	12.5
Repetitive practice	16.5	23.3	8.7	12.5
Experimental practice	15.2	0.4	12.7	18.7
Silent practice	5.2	24.1	29.3	6.2

Terry: Creative-narrative perspective

Unique in Terry’s practice was her use of extramusical imagery in relation to the title of the piece. She used various narrative/associative strategies, through which she developed a background story to the piece (see Table 1; Appendix 2). Her analytical and aural observations were often interconnected, with visual, emotional, and movement imagery and verbal associations. In addition to illustrating the music, she also used narrative/associative strategies for expressive interpretation and creating mnemonics. We characterized Terry’s practice perspective as creative-narrative.

Initial silent reading

First, immediately after seeing the title, Terry reported “thinking of where [in the music] the ‘moo’s might be and where this animal is swinging its tail.” From the video recording of her eye movements it would appear that she viewed the title three times before proceeding to read the piece from beginning to the end. While reading, she reported imagining the rhythms “inside of my mind silently,” but confessed also being “a person who doesn’t start to sing [pitches in my mind] when seeing the score. That’s why the picture [of the score] gives [me] some ideas and presentiments.” Terry’s conception of the musical structure was a narrativized one, revealing some visualization and observations on narrative elements:

It started peacefully with such small [movements] close by. And then, when the left hand comes along [. . .] there will be events [. . .] [events of] what might be happening on the grass. It seemed like a cow on the grass, not in a cowshed.

First Practice on the Piano

Immediately afterwards, Terry recalled “just playing” the piece through “a few times,” likely for getting an aural image of the piece. The practice data revealed that Terry spent 45.5% of her time on linear practice, playing the piece twice through and once from bar 9. In addition, she focused on certain bars using fragmented and repetitive practice (Figure 2; Table 2) that may relate to the use of problem-solving strategies such as “checking, kind of reading what notes there are” and “planning hand divisions.” Of all the participants, Terry’s experimental practice most often involved leaving the keys depressed while she seemed to be listening or reflecting.

Figure 2.

Different types of practice types used by Terry in each bar. In addition to linear playing, she focused on certain bars using different types of practice, also often leaving the keys depressed as if while listening or reflecting.

Terry’s observations on the music’s atmosphere, during the interview, were interconnected with visualizing the scenery. She described the music as a “peaceful state of mind” where the “cow is on the pasture; even though there might be other cows, too, they don’t mess around.” She associated the 7/4 meter with the cow’s movement: “In the quadruple measures it’s pretty clearly [movement] from one side to another” and “the 3/4 bar is like swinging of the tail.” The music had emotional associations for Terry, “sounding as if the cow is having some worries.” Some of her detailed associations were related to local musical elements. For example, she pondered whether the A flat on the third system “might be the cow’s head that is thinking?” At bar 13, Terry experienced the 32nd (demisemiquaver) notes as “a herding call” and stated: “Now it’s coming [. . .]: the milker!” At bar 19 she made an association with the cow’s bellow, wondering: “What kind of a voice does a cow have?” At this bar, Terry’s repetitive and fragmented practice seems to relate to reflections regarding temporal expressivity (“how slowly should I take it?”).

First week’s practice

During the first week, Terry extended her associations to expressive dynamics: “There are only quiet nuances. [. . .] [Hence], this is a summer field—no storm clouds anywhere near.” She stated seeing images “just like a film” and remembered “already as a child and a teenager [. . .] always reading [. . .] and always having clear images of how a certain character in a book looks like and in what kind of a landscape.” Similarly, Terry described the cow’s appearance, activities and the surroundings:

Lovely green grass on the field [. . .] about fifteen twenty centimeters [long]. [. . .] The landscape is not flat, but it’s not an Alpine landscape either [. . .] All the time I have seen a cow with chubby loins. I’ve seen it brown and white. [. . .] Sometimes I’ve seen it raising her head and smelling, the tongue going first to one nostril and then to the other one.

At this point, Terry started creating verbal mnemonics: “I have decided to create lyrics” to the music, “just inside my head—to assist my memory.” Terry noted that by creating lyrics, she also “marked the keys with something visual.” As an example, she played from bar 9: “I think I [i.e., the cow] will close my eyes [. . .]. And here, the eyelashes are closing [plays the left hand at the beginning of bar 10]” (Figure 1). Another example of the multiplicity of Terry’s associative processes is the “herding call passage” at bar 13, where she had experienced the symmetric hand position as a visual “landscape” at the keyboard: “F sharp as a central note at both edges, and the thumbs play F and G [. . .] This is the green [plays F and G], and here are the fences [plays the registrally flanking notes F sharp and G flat].”

Second week’s practice

In the interview after the concert, Terry mentioned focusing on “what I should express with it [the piece]?” In the performance, she reported thinking of “the sound” and relying on the lyrics: “I was just hoping that my mouth wouldn’t move along.” Despite Terry’s earlier plans, she had created lyrics only to some key passages or key pitches. For example, she had marked the note A flat in the beginning verbally (“This is the ‘moo’”) and the note F sharp visually by drawing a sun above it in the score. Terry elucidates the lyrics created during the second week:

“Oh yes, the sun is shining [. . .] I eat, I eat.” [. . .] And while it eats we get to the third system, where the first bar goes: “My rumen is full again. Moo.” We are back at A flat [. . .] And then, in the end of the third system, the left hand: “Ruminate. Am-moo. Ruminate.”

Finley: Temporal-normative perspective

Finley’s meticulous strategies included correct implementation of all details with particular focus on the rhythm, segment-wise practice, immediate solving of practical problems such as fingerings, and immediate memorizing (see Table 1; Appendix 2). In addition to precise treatment of rhythm, Finley showed interest in the music’s direction by listening to melodic movement and pitch interval relations. She described the piece using expressions related to rhythmic character. She consistently seemed to apply strategies that were reportedly based on her experiences of learning and memorizing contemporary music. For these reasons, we chose to characterize this perspective as temporal-normative.

Initial silent reading

From the score, Finley “immediately observed that it was so atonal that I decided just to look at the big [picture] and maybe more at the rhythmic structure.” Her analytical focus on temporal details can be seen in how she describes the piece from memory after silent reading:

The time signature was 7/8 [7/4]. I think the tempo marking for a half note was 52. There was pretty much chromaticism and polyrhythms, a lot of tied eighth notes in the structure. [. . .] Chords in the bass register. Pretty tight looking texture. [. . .] Somewhat rapid rhythmic items, like written-out ornaments.

Finley reported “first just skimming from beginning to the end really fast to get an overview of the material: ‘What’s in there?’ [. . .] Then I started to go through it more systematically.” This was supported by the video recording of her eye movements. First, she spent 13 seconds skimming both pages, and continued by reading slowly (“a bit more detailed”) through the first page. At bars 10–11 Finley showed interest in rhythm as she appeared to spend 28 successive seconds using motor imagery strategies: “Where the left hand joins in, I was kind of trying to play the rhythm [. . .] with both hands [to find out] where the beats are.”

First practice on the piano

Finley’s first practice was unique and systematic in terms of segmentation. Her linear practice was never on more than four consecutive bars. Finley described “taking a small segment that I might be able to study effectively and well,” and played only bars 1 to 10, using systematically similar practice (Figure 3). This involved moderately large amounts of repetitive and fragmented practice (together occupying 46.3 % of her total practice time), and hardly any experimental practice (Table 2). During practice she reported problem-solving strategies such as planning and “testing the fingerings—whether the transitions work well,” but also “trying to follow the dynamic markings.” Meanwhile, she pursued the melodic movement by “singing it in my mind [. . .]. That way it stays in that [aural] memory.” These strategies seemed motivated by Finley’s experience of contemporary music: “The more modern the music is, the better my memory works if I have done exact and precise work from the beginning.” At this point, Finley described the music using terms related to abstract movement (not related to specific character): “swirling, wobbling in a fairly narrow range.” She also seemed to aim at temporal understanding and effective learning during her silent practice activities. In this phase she “wrote fingerings” and other “attention marks,” but at around bar 10 Finley took a long break to study the rhythm in detail, trying to make sense of “which notes are tied, and which ones should be played.” In this area, Finley’s detailed rhythmic work is reflected in her practice with an instrument as almost exclusively fragmented or repetitive practice (Figure 3).

Figure 3.

Different types of practice types used by Finley in each bar. They demonstrate a segmented and systematic approach. She practiced only bars 1–10. The total amount of fragmented and repetitive practice was the largest of all the participants (46.3%), and the amount of linear practice was smallest (29.3%).

First week’s practice

During the first week, Finley reported trying to be “really accurate” with rhythms, and assisted this temporal work by using the metronome, “systematically accelerat[ing] the tempo.” She experienced the piece to “include interesting rhythmic playfulness,” and associated the repeating “swinging” motif in the beginning with a “perpetuum mobile.”

Finley’s aural strategies were directed toward melodic movement, and she reported “singing in the mind.” From her teacher Finley had learned to “listen to the pitch intervals like a singer,” and stated that “even though the music might not be the most beautiful in the world [. . .], there are always relationships.” In Finley’s experience, interpretational work “comes through inner singing,” but reported also “thinking of [dynamic] nuances and trying to implement the markings.” She noted: “Despite that it has been marked as quiet and espressivo, the layers need to be heard.”

Second week’s practice

After the concert, Finley stated that “processing the rhythm” was essential in her practice: “You should be able to play it in [correct] tempo and in [correct] rhythm—that you really know the rhythms.” She reported having struggled with “whether it [the rhythm] can be free, or should it be very rhythmical,” finally deciding to approach the piece “authentically,” meaning precise rhythms. She experienced the music as “intellectually composed,” referring especially to “the varying rhythmic aspects.” In the end of the process, Finley’s practice appeared more holistic: “I started thinking about larger wholes,” “always thinking of where the music is going to.”

Kim: Aural-atmospheric perspective

Characteristic of Kim’s practice was her interest in sound quality and the aural atmosphere of the music. Kim seemed to be interested in both local sound effects as well as general atmosphere. Her thematic and harmonic observations were partly mediated by aural processes and interconnected with atmospheric impressions. For these reasons (see Table 1; Appendix 2), we chose to characterize Kim’s perspective as aural-atmospheric.

Initial silent reading

Kim reported possessing absolute pitch. From her descriptions it appears that even while reading silently she was able to gain an aural overview of the music and its atmosphere. She reported first “playing it through in my mind,” and while doing this, “listening to the voices the best I can,” although at passages that were “difficult to read I was just leaping over to get a holistic picture of the music.” The video recording of her eye movements suggests that after quickly skimming the beginning of the piece, Kim did indeed read the whole piece linearly from beginning to the end, fluently and without interruption. While the piece “didn’t sound like a cow” in Kim’s opinion, her analytical observations were coupled with descriptions of the sonic atmosphere:

[It sounded] a bit modern of course, but there was this exciting atmosphere [. . .]. No major or minor things [harmonies], but quite a lot of tritones and such, so it gives a bit unclear [sound]. [. . .] Mystic or a bit exciting: a bit hazy and foggy.

First practice on the piano

Immediately afterwards, Kim reported “trying, of course, whether I was able to play this through.” However, “I did not want to spend that much time on the two first systems.” Instead, Kim emphasized “the passages where there is most stuff—the second half, as well as the rhythms with some timing issues between the voices.” Focus on bars 10–16 is visible in Kim’s practice data (Figure 4). Of all the participants Kim, as an absolute pitch possessor, may have been least dependent on the instrument: almost 30% of her practice was labeled “silent practice” (Figure 4; Table 2), during which she reported “analyzing or listening in my mind.” Her silent practice apparently also involved reflecting atmospheric effects such as dynamics and articulation, which “I wasn’t necessarily playing. I may have just been thinking of them.”

Figure 4.

Different types of practice types used by Kim in each bar. Some of her repetitive, experimental, and silent practice was identified as connecting with aural-atmospheric strategies.

Overall, Kim seemed interested in the composer’s expressive ideas: “I tried to think of what [the composer] has been after in a musical sense.” Instead of “thinking of performance interpretation yet, it was more like making the music clear to myself [. . .]—what the atmosphere is and what the touch that is needed.” In her practice data, the moderately high amount of linear and silent practice (Table 2) compared to fragmented and repetitive practice (together, 20.9%) might indicate Kim’s reflective work on the general atmosphere. Her work on articulation is seen in the practice data as repetitive (and experimental) practice in bar 19, at least (Figure 4): here, she reported “greatly wondering [. . .] the F sharp bass—there is a slur and this long note should continue to sound. Should it be taken again as a mute note because there is a line and an accent?” Kim described the piece as “pretty exciting,” continuing: “Perhaps it’s the dynamics of the piece: it’s in a pianissimo world. [. . .] There are colors or sort of effects” (referring to the tetrachords on p. 2 as “signals”).

First week’s practice

During the first week, Kim reported “pretty much just playing it through and trying to find the essence of the piece.” Kim reported that central to her first week’s practice was “music analysis, but also getting the image of the sound in my ear, [. . .] creating an atmosphere, and understanding what the composer has been after.” She had an impression that the sound “should not be too heavy, but rather light and appealing.” When asked to tell the researcher about the piece, she said, “It has a great atmosphere and nice chords—a bit like bells.” Kim’s own idea of the general atmosphere conflicted with the metronome marking: “I was surprised that it [the metronome marking] is quite fast. The feeling I got from the score was dreamier.” Although Kim described the music in somewhat visual terms as a “momentary image,” “impression,” or “landscape,” this landscape—unlike Terry’s previously—was a specifically aural one, “a sound image.” Instead, Kim reported “in general not illustrating music. [. . .] Music is the language through which it [i.e., the piece] speaks to me.” However, she described this aural image metaphorically:

It’s a bit misterioso [. . .] pretty soft, not a conflict. It’s like a foggy morning or such. [. . .] The atmosphere is fairly pastoral. [. . .] Not agonizing, just a bit unclear landscape, like a forest in dew, sounds heard here and there, far away.

Second week’s practice

During the second week, Kim mostly focused on memorizing. She decided to play the piece lyrically, acting upon her own interpretation regarding slower tempo. Kim’s aural-atmospheric perspective was confirmed in the statement that the music “introduced colors, feelings” to her.

Morgan: Compositional-analytic perspective

Morgan’s practice was characterized by analytical and aural examination of motifs, pitch structure, and the sound of the structure. Along the way, she was reflecting the composer’s intentions (see Table 1; Appendix 2). Because of this combination of strategies, her practice perspective was termed compositional-analytic.

Initial silent reading

While reading the piece silently, Morgan assumed that it might be based on a 12-tone row. She claimed to be able to imagine the music aurally “quite easily,” and said the piece “sounded pretty pleasant and peaceful.” She reported first “going it once through”—a process confirmed by the video recording of her eye movements—and then examining the pitch structure: “I was just looking at the first system [. . .] thinking of the pitches—whether there is a [12-tone] row or not.” The eye-movement video suggested that she did indeed return to read the beginning of the piece, for a total of 27 seconds.

Morgan showed interest in how the idea of a cow was manifested in the composition, thinking: “Does it sound like a cow, does it look like a cow?” The “lack of big dynamic changes” evoked associations of the cow’s condition: “Clearly, the cow is pretty healthy and not getting any seizures or such [laughing].” Morgan reflected the composer’s intentions as she reported thinking of whether the “meandering and variable” rhythmic materials “have a meaning,” concluding that with “many binding slurs crossing the bar lines,” the composer has “wanted to break regularity.”

First practice on the piano

During Morgan’s first practice, she “just wanted to play [the piece] quite a lot because it has a nice sound.” As can be seen in Table 2, 50.2% of her practice time was coded as linear, and only 6.2% as silent. Of all the participants she used the largest amount of experimental playing (almost 19%), often, apparently, so as to listen to the sound structure and harmonies. For example, at bar 15, she appeared to have listened to the “pretty good sounding” chord she “associated with A flat major” (Figure 5).

Figure 5.

Different types of practice types used by Morgan in each bar. Some of her experimental, fragmented, and repetitive playing was related to studying the pitch structure of the piece and “savoring” the harmonies.

First, Morgan “started playing [the piece] through in order to see whether its idea corresponds to what I was thinking while [silently] reading.” She also reported “perhaps stopping at some passages just to play them.” The practice data confirmed that she first played the piece through, only stopping for experimental, fragmented and repetitive playing at bars 13–16. After playing the piece through once, she “was looking at it with a more analytical eye, playing at the same time.” Indeed, she started again from the beginning, soon returning to experimental practice of the opening bars for 25 seconds (Figure 5). At this point, Morgan reported “playing the notes in the first system and leaving [the keys] depressed,” becoming convinced that “in the beginning, [the 12-tone row] is pretty freely presented.” She also reported reflecting the composer’s intentions: “The repeating motif with B flat and E [. . .] Does it have a meaning?”

Morgan thought that the music “might be mimicking a cow’s voice a bit,” and interestingly gave the same example as Terry—bar 19—which in Morgan’s opinion “is like low bellowing that they [cows] sometimes make.” Her “savoring” of this pattern may be seen as the most repetitive playing of all the bars (Figure 5).

First week’s practice

Because Morgan’s practice time was limited she used silent reading strategies on a train, “writing down fingerings,” and “playing in mind.” During the week, she reported playing on an instrument only once for “checking the [silently planned] fingerings and listening to the music.” Morgan also reported silently analyzing the pitch structure “playing the keyboard [in imagery], trying to memorize how the pitch intervals behave, and how many times they repeat.” She studied “specifically just the pitches, not the rhythms nearly at all,” which “is something I often do—especially with atonal music that needs to be memorized.” Morgan organized melodic pitches creatively as harmonies: “The [12-tone] row forms harmonies—bulks that I experience as chords.” This is how Morgan demonstrated this harmonic structuring of the material on the piano:

I didn’t think of the music that much as bars, but rather as fragments related to these chords [. . .] This is one such fragment [Morgan plays the first nine pitches], then the next one here [plays the 17 following pitches] and the third [plays the 11 following pitches], and fourth [plays the last 12 of the 2nd system] (Figure 1).

At this point, Morgan seemed most interested in how the music sounds, stating: “What is interesting here is the sonore.” For Morgan, a good sound depended on “how the texture has been arranged,” and she acknowledged the current piece is not one of those compositions “written on paper without thinking of the sound. [. . .] Here, in my opinion, it has been well thought out.” She also observed that the overall form “includes a very clear culmination. [. . .] Then, the same motif is repeated many times, and after that there is kind of a small coda. That structures the piece well.”

Second week’s practice

During the second week, Morgan reported learning the piece “systematically with an instrument,” and focused on the interpretation and markings in the texture. She also analytically paid attention, for example, to how the composer “used the chords in a certain light manner [. . .] larger chords are in a higher register.” Morgan, a composer herself, saw the sounding pitch structures as the most essential aspect of the music: “Even though there are interesting rhythms, they are not the main point.”

Conclusions

Study I confirms that expert musicians can use the same practice strategies, even at the same stages of practice, but still have different cognitive orientations while practicing the same piece of music (cf. Hallam, 1995; Hultberg, 2008). One common strategy involved making a visual and temporal overview of the piece while reading it silently for the first time (cf. Héroux, 2018). The participants also used embodied strategies throughout practice in much the same way, and interpretive and memory strategies to a greater extent in the later stages of practice. All the participants used temporal, analytical, aural, and interpretive strategies. However, we identified for all four participants an individual practice perspective involving focus on different types of information and indicating somewhat different conceptions of the piece. Terry worked with extramusical information, approaching the piece as a narrative. Linear and rhythmical aspects of the piece were central to Finley’s approach as she emphasized articulating its temporal movement and form. Kim focused on aural information and heard the piece in terms of sonic atmosphere, whereas Morgan attended to its structure, aiming to understand the compositional ideas underlying the piece. These results show how even a short piece of music—as a complex amalgam of sounding and ideational elements—can afford a variety of cognitive approaches.

In accordance with the findings of Dos Santos and Hentschke (2011), personal skills, interests, values, and educational experiences seemed to guide the musicians’ orientation to their repertoire. Terry’s creative-narrative and Kim’s aural-atmospheric perspectives could perhaps be characterized as more intuitive approaches focusing on “artistic image” of the music (see Héroux & Fortier, 2014; Neuhaus, 1993). These perspectives seemed to have at least some basis in the pianists’ skills and tendencies in using imagery. Terry created a background narrative to the music but also seemed to have harnessed her reportedly vivid and creative imagery to serve practical purposes such as memorizing. Kim, in turn, appeared to have advanced aural skills but no preference for illustrating music in visual terms. She focused on listening, reflecting, and delivering an aural-atmospheric message. The other two perspectives—Finley’s temporal-normative and Morgan’s compositional-analytic—could accordingly be understood in terms of an analytical approach and interest in the “formal image” (see Héroux & Fortier, 2014) of the piece. These two perspectives seemed affected by educational and professional history. Finley mentioned attitudes inherited from her previous teachers, and both pianists reported habits and experience in memorizing contemporary music. Morgan’s skills in thinking about compositional design and composers’ intentions were potentially affected by her own compositional activities. Thus, it may be that musicians see a musical piece in the same way that people in general view the world around them—through their own individual properties, in accordance with the principle “like knows like” (Watson, 1995, p. 10).

Study II

Listening experiment: Expert evaluations of concert performances

Given the results of Study I, our second research question could now be rephrased: “Can musicians’ individual practice perspectives, applied during practice, be detected by experienced listeners when they listen to the musicians’ final performances?” To answer this question, we created an online listening study, inviting professional piano teachers to participate as expert listeners.

Method

Participants

We sent an email invitation to piano pedagogues at music institutes in different areas of the country. Before sending the invitations, the information sheet concerning the online study and data protection were reviewed by the data protection officer of the University of Jyväskylä. The mean age of the 34 participants (26 females, 8 males) who replied was 47.6 years (SD = 8.6 [range 31–61]). They had spent a mean of 37.5 years in goal-oriented piano practice (SD = 9.4), worked as piano teachers for a mean of 22.8 years (SD = 10.5), and 18.3 years as performing artists (SD = 12.8).

Musical materials

The stimuli consisted of audio-recordings of the four performances by the participants in Study I, as well as an additional performance of the same piece played by a professional pianist to serve as a practice trial. The duration of this performance was the mean duration of the four concert performances (see Wöllner, 2013).

Procedure

A survey was created and run on an internet platform (Webropol). The performances were evaluated using a 12-item questionnaire in which each of the four practice perspectives identified in Study I was represented by three items (see Appendix 3). Two of these represented the respective participants’ characteristic practice strategies while one represented her mode of experiencing the piece. These items were derived from the interview data gathered in the middle and concert stages of the study.

First, respondents to the survey undertook a single practice trial to familiarize themselves with the task and the piece. Then they undertook the four main trials. Their task was to listen to each performance as many times as they wished, judging on five-point scales (1–5) to what extent they thought each questionnaire item characterized the performer’s approach to practice. The order of the four performances was counterbalanced, and the order of the questionnaire items was randomized.

Data analysis

The data were analyzed using R statistical software (R core team, 2019), using the “psych” package (Revelle, 2019) to carry out principal component analysis. The component scores were subjected to linear mixed-effects model analysis in the “lme4” package (Bates et al., 2015), and estimated marginal means were produced using the “emmeans” package (Lenth et al., 2019).

Results

First, we wanted to find out which questionnaire items would receive significantly different ratings for the four performances. None of the questionnaire items was normally distributed (Shapiro-Wilks tests, all ps < .001), so we used nonparametric Kruskal-Wallis tests. Only two items received significantly different ratings: “experiencing the piece as a momentary image with mystic atmosphere and bells chiming far away” (χ²[3] = 14.17, p = .003) and “practicing authentic, precise rhythmic treatment” (χ²[3] = .95, p = .047). The first of these two items was derived from Kim’s practice, but the mean rating for her performance (3.09) was not the highest, as the mean ratings for the other performances were 3.74 (Terry), 3.27 (Morgan), and 2.59 (Finley). The other item was derived from Finley’s practice, but once again the mean rating for her performance (3.47) was not the highest, as the mean ratings for the other performances were 3.24 (Terry), 3.82 (Kim), and 3.18 (Morgan).

Next we asked if the differences between the ratings, for these items, could be explained by the pianists’ use of different practice strategies. At the crudest level of analysis, the mean ratings for the “momentary image” item (created to represent Aural strategies) showed a near-significant Pearson correlation with the number of coded instances of the use of Narrative/associative strategies (r = 0.95, n = 4, p = .053), but with no other strategic families (all ps > .1). The “rhythmic treatement” item showed no significant correlations with any of the strategic families (all ps > .1).

To explore the influence of pianists’ practice strategies on listeners’ assumptions regarding the practice, we carried out a principal component analysis of the listeners’ ratings, extracting three components with eigenvalues over 1 using varimax rotation (see Appendix 3). The first, largest, principal component loaded heavily on aurally centered practice strategies such as “searching for the sonic atmosphere of the piece,” and generally covered all of the “experience” items of the questionnaire. We call this component Atmosphericity. The second component appeared as Extramusicality, loading on the practice strategies derived from Terry’s creative-narrative practices. The third component, Rhythmicity, was centered on the strategy of “practicing authentic, precise rhythmic treatment.”

Using the component scores from the principal component analysis (normalized between 0 and 1), we subjected each principal component to linear mixed-effects model analyses, with listener as the random effect and the coded strategies as predictors. For each response variable (i.e., principal component score), we first built a “null” model with only the expert listeners as a random effect, and proceeded by likelihood-ratio tests, considering the predictors one by one. In the case of Extramusicality, there were no statistically significant fixed effects. For Atmosphericity, the chosen model included only a fixed effect of assumed narrative/associative strategies (χ²[1] = 5.06, p = .024). For Rhythmicity, our model similarly included merely narrative/associative strategies (χ²[1] = 10.71, p = .001). No other fixed effects (or interactions) could be added by further likelihood-ratio tests in either case. The regression coefficients are shown in Table 3, and the predicted values from the model are shown in Figure 6. Our models predicted that if a pianist uses narrative/associative strategies, listeners would be more likely to believe, from the heard performance, that the pianist had taken an atmospheric orientation approach to practice, and less likely to believe they had taken a rhythmic orientation approach. However, the proportion of variance explained by the fixed effects was low (see Table 3). It is relevant to note that the use of narrative/associative strategies did not seem to support listeners’ attributions of an extramusical orientation to the practice process.

Table 3.

Linear mixed-effects models for the principal component scores of listeners’ judgments.

	Atmosphericity (marginal R² = 0.05, conditional R² = 0.15)
	Estimate	Std. Error	df	t value	P (> \|t\|)
(Intercept)	.4688	0.0294	9.56	15.946	< .001***
Narrative/associative strategies	.0063	0.0022	10.21	2.884	.005**
		Rhythmicity (marginal ^R2 = 0.06, conditional R2 = 0.29)
	Estimate	Std. Error	df	t value	P (> \|t\|)
(Intercept)	0.6250	0.0264	75.15	23.695	< .001***
Narrative/associative strategies	−0.0056	0.0017	102.09	−3.241	.002**

Significance levels: *** p < .001; ** p < .01

Fig. 6.

Predicted values (and 95% CIs) of Atmosphericity and Rhythmicity for various levels of narrative/associative practice. In both panels, the four pianists appear in order of increasing use of narrative/associative strategies: Finley, Kim, Morgan, and Terry.

Conclusions

The main result of Study II was that experienced listeners could not detect the the four pianists’ characteristic practice perspectives from their performances. For instance, although three different narrative/associative strategies formed a consistent dimension in the listener judgments, they were not especially often attributed to Terry from whose practice they had been derived. Perhaps this is not surprising: even though individual musicians may study the pieces through temporal, atmospheric, extramusical, or compositional elements, for example, their cognitive orientations represent only part of what affects the final performance. As suggested by Juslin (2019; Juslin & Laukka, 2003), performance expression in classical music is multidimensional, partly deriving from sources outside of voluntary control. Although musicians’ characteristic practice perspectives might not be heard in their performances by listeners, our results suggest that certain practice perspectives might indirectly affect the perceived qualities of a performance. To a small extent, a focus on narrative/associative strategies seemed to increase the likelihood of our expert listeners assuming that the pianists had attended to the sonic atmosphere, and decrease the likelihood of assumptions regarding focus on rhythmic precision while practicing. A possible interpretation would be that even if relying on extramusical associations may direct the musician’s own focus in her performance (e.g., away from rhythmic details), what is primarily conveyed to the listeners are just the qualities they hear in the music as it is performed.

General discussion

In this research, we studied skilled pianists’ strategies while they practiced the same post-tonal piece over the course of two weeks. We also investigated whether experienced listeners could detect the pianists’ characteristic approaches to practice from their final performances.

In Study I, the four participants used similar strategies, some at the same stages of practice, as is common for expert musicians (e.g., Hallam 1995; Hallam & Bautista, 2018). However, we also identified a characteristic practice perspective for each pianist that involved consistent families of strategies focusing in particular on certain types of musical or extramusical information. Terry created a narrative of the music using extramusical imagery, Finley approached the piece through temporal thinking, Kim through local and global sonic atmospheres, and Morgan studied the way the music had been composed. Our case studies show that expert classical musicians may differ extensively in the ways they conceptualize music. Our results suggest that musicians use analytical and intuitive approaches to different extents (cf. Héroux, 2018), and may be drawn more to one or the other. We tentatively traced some aspects of the pianists’ practice perspectives, through their self-reports, to their reported imagery skills and tendencies, practice habits, and the skills and preferences they had developed through their professional activities and education (see also Dos Santos & Hentschke, 2011; Héroux, 2018). An obvious limitation of the study was that it only involved four participants. Likewise, our selected piece “La vache,” with its programmatic title and its rhythmical features, may have affected some of the perspectives intuitively taken by the pianists. In future studies, additional practice perspectives and their potential origins could be identified, and more generalizability achieved, by involving more participants, a wider selection of music, and different musical genres.

In Study II, expert listeners were found not to be able to associate heard piano performances with the pianists’ actual practice strategies and modes of experiencing the piece. Principal component analysis revealed, however, that expert listeners were able to attribute features such as “extramusicality” to performances in a cohesive way. The finding that it is difficult to associate performances with their actual practice histories might be accounted for in three ways. First, musicians’ practice approaches are multidimensional in terms of the strategies they use, and different musicians may share many similar strategies. Second, performance expression is affected not only by practice strategies but also by mechanisms beyond the performer’s control (see Juslin, 2019; Juslin & Laukka, 2003). Third, some practice strategies, such as those we termed narrative/associative ones, may not have obvious musical results from which they could reliably be detected. For these reasons, identifying practice strategies would require a rather intuitive approach, and might indeed be close to impossible using an ecologically valid method such as ours that allows musicians to approach the piece in their preferred ways. We cannot overlook the possibility that experiments in which individual musicians would be recommended to focus on certain musical parameters in their practice might result in more accurate listener ratings. In future studies, computational analysis of performances in conjunction with broader samples of practice might help to reveal connections between practice perspectives and “musical fingerprints” of various kinds (see Bernays & Traube, 2014; Van Vugt et al., 2013). Beyond the strict identification of practice strategies, the present results also point toward the potentially fruitful question as to whether the extramusical imagery used by musicians might affect the mood of the music as perceived by the listener.

In our study, we considered initial silent reading as an opportunity to compare participants’ ways of making observations on the score and starting to develop mental representations of the piece. This did indeed reveal their interests in different kinds of musical and extramusical information that continued to characterize the later stages of practice. In future, it would be worth investigating whether “silent reading styles” (Penttinen et al., 2013) alone might offer a tool for assessing practice perspectives, perhaps using self-assessment forms to gather complementary evidence. It should be noted, though, that the pianists in the current study were selected on the basis of their excellent memorizing skills. It seems likely that less proficient memorizers might rely even more heavily on musical notation as scaffolding for their practice processes.

It might be reasonable to ask why analyzing the performer’s mindset during practice should be relevant for the listener. What is the role and the value of the kinds of practice perspectives discovered here? We suggest that musicians may be drawn to specific types of information in music, processing musical pieces through their own strengths or preferences. Some of these special skills and tendencies might be taken for granted by the individual musicians themselves. To the extent that self-knowledge is crucial for constructive and conscious personal and professional development (e.g., Elia et al., 2016; Wilson, 2009), encouraging musicians to learn about their own working processes might help them understand their own “musical personalities,” and broaden their views beyond customary perspectives. Given that professional musicians often work on solo repertoire by themselves (Jørgensen, 2004), the ability to use a wide spectrum of practice strategies might increase their ability to self-regulate.

When musicians practice and rehearse, they put their personal style on the line, laying the foundation for listeners’ judgments concerning the individuality of the performances. Yet the musicians’ thoughts during practice may be unreachable and even irrelevant to the listener, for whom the main focus may often be the quality or uniqueness of the performance. Taking this point of view, we suggest that classical musicians’ practice perspectives might primarily be significant as private meaning-making devices that enhance performers’ intrinsic motivation within the musical domain, reflecting their personal artistic engagement and their ways of building relationships with musical works.

Footnotes

Appendix

Appendix 3.

Principal component analysis of the ratings in Study II.

Pianist of origin	Suggested practice strategy (or experience):^a (“To what extent has the performer during their practice been. . .”)	PC1	PC2	PC3	Communality
Terry	composing lyrics related to the cow to the musical motifs.		0.88		0.88
	visualizing a chubby cow on a sunny summer pasture.		0.87		0.86
	experiencing the overall structure as a narrative.	0.69	0.42		0.66
Finley	practicing authentic, precise rhythmic treatment.			0.91	0.85
	trying to listen to pitch relations (intervals) like a singer.	0.65			0.58
	experiencing that layers in the music should be heard in the performance.	0.78			0.74
Kim	searching for the sonic atmosphere of the piece.	0.89			0.84
	investigating the timbral effects of the composer.	0.87			0.84
	experiencing the piece as a momentary image with mystic atmosphere and bells chiming far away.	0.76			0.70
Morgan	listening to the piece’s pitch structure and harmonies formed by pitches.	0.80			0.77
	parsing the overall structure of the piece in mind into different parts.			0.52	0.62
	experiencing that the composer’s design in using pitch registers make the piece sound well.	0.84			0.83
	Sum of squared loadings	5.367	2.320	1.445
	Proportion of variance	0.45	0.19	0.12
	Cumulative variance	0.45	0.64	0.76

Evaluated on a scale from 1 (“very little or not at all”) through 5 (“very much”).

PC1 = Atmosphericity; PC2 = Extramusicality; PC3 = Rhythmicity.

Acknowledgements

The authors are grateful for the two reviewers for their valuable comments which helped to improve the paper. We also want to thank Jane Ginsborg for editorial work, Marc Thompson and Marjaana Puurtinen for all help and support, Birgitta Burger and Emma Allingham for help with Dikablis, and Alvaro Chang Arana for assistance in piloting the study. Most of all we want to thank our four expert pianists of Study I for their commitment, and the expert listeners of Study II.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was funded by Academy of Finland, Project number 275929, project name: ”Reading music: Eye movements and the development of expertise,” The Finnish Concordia Fund, and The University of Jyväskylä.

ORCID iD

Nina Johanna Loimusalo

References

Altenmüller

Shinichi

(2016). Planning and performance. In Hallam

Cross

Thaut

(Eds.), The Oxford handbook of music psychology (2nd ed., pp. 529–545). Oxford University Press.

Bates

Mächler

Bolker

Walker

(2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67(1), 1–50. https://arxiv.org/abs/1406.5823

Baxter

Jack

(2008). Qualitative case study methodology: Study design and implementation for novice researchers. The Qualitative Report, 13(4), 544–559. https://nsuworks.nova.edu/tqr/vol13/iss4/2/

Bernays

Traube

(2014). Investigating pianists’ individuality in the performance of five timbral nuances through patterns of articulation, touch, dynamics, and pedaling. Frontiers in Psychology, 5(157), 1–19. https://doi.org/10.1177/1029864915572813

Chaffin

Demos

A. P.

Logan

(2016). Performing from memory. In Hallam

Cross

Thaut

(Eds.), The Oxford handbook of music psychology (2nd ed., pp. 559–595). Oxford University Press.

Chaffin

Imreh

Lemieux

Chen

(2003). “Seeing the big picture”: Piano practice as expert problem solving. Music Perception, 20(4), 465–490. https://doi.org/10.1525/mp.2003.20.4.465

Creswell

J. W.

Creswell

J. D.

(2018). Research design: Qualitative, quantitative, and mixed methods approaches (5th ed.). SAGE.

Dos Santos

R. A. T.

Hentschke

(2011). Praxis and poiesis in piano repertoire preparation. Music Education Research, 13(3), 273–292. https://doi.org/10.1080/14613808.2011.603042

Elia

Aprá

Verhovez

Crupi

(2016). “First, know thyself”: Cognition and error in medicine. Acta Diabetologica, 53, 169–175. https://doi.org/10.1007/s00592-015-0762-8

10.

Gabrielsson

Juslin

P. N.

(1996). Emotional expression in music performance: Between the performer’s intention and the listener’s experience. Psychology of Music, 24(1), 68–91. https://doi.org/10.1177/0305735696241007

11.

Ginsborg

Chaffin

Demos

A. P.

(2012). Different roles for prepared and spontaneous thoughts: A practice-based study of musical performance from memory. Journal of Interdisciplinary Music Studies, 6(2), 201–231. https://doi.org/10.4407/jims.2014.02.005

12.

Hallam

(1995). Professional musicians’ approaches to the learning and interpretation of music. Psychology of Music, 23(2), 111–128. https://doi.org/10.1177/0305735695232001

13.

Hallam

Bautista

(2018). Processes of instrumental learning: The development of musical expertise. In McPherson

Welsch

(Eds.), The handbook of music education (2nd ed., pp. 108–125). Oxford University Press.

14.

Héroux

(2016). Understanding the creative process in the shaping of an interpretation by expert musicians: Two case studies. Musicae Scientiae, 20(3), 304–324. https://doi.org/10.1177/1029864916634422

15.

Héroux

(2018). Creative processes in the shaping of a musical interpretation: A study of nine professional musicians. Frontiers in Psychology, 9 (665), 1–17. https://doi.org/10.3389/fpsyg.2018.00665

16.

Héroux

Fortier

M.-S.

(2014). Expérimentation d’une nouvelle méthodologie pour expliciter le processus de création d’une interprétation musicale. Les cahiers de la Société Québécoise de recherche en musique, 15(1), 67–79. https://doi.org/10.7202/1033796ar

17.

Hultberg

(2000). The printed score as a mediator of musical meaning: Approaches to music notation in Western tradition. Malmö Academy of Music, Lund University.

18.

Hultberg

(2008). Instrumental students’ strategies for finding interpretations: Complexity and individual variety. Psychology of Music, 36(1), 7–23. https://doi.org/10.1177/0305735607079719

19.

Jolivet

(1986). La vache. In Khitrouk

(Ed.), Musique française: Pièces choisies (pp. 12–13). Muzyka.

20.

Juslin

P. N.

(2003). Five facets of musical expression: A psychologist’s perspective on music performance. Psychology of Music, 31(3), 273–302. https://doi.org/10.1177/03057356030313003

21.

Juslin

P. N.

Friberg

Bresin

(2001). Toward a computational model of expression in music performance: The GERM model. Musicae Scientiae, 5(1), 63–122. https://doi.org/10.1177/10298649020050S104

22.

Juslin

(2019). Musical emotions explained: Unlocking the secrets of musical affect. Oxford University Press.

23.

Juslin

Laukka

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

24.

Jørgensen

(2004). Strategies for individual practice. In Williamon

(Ed.), Musical excellence: Strategies and techniques to enhance performance (pp. 85–104). Oxford University Press.

25.

Keller

P. E.

(2012). Mental imagery in music performance: Underlying mechanisms and potential benefits. Annals of the New York Academy of Sciences, 1252, 206–213. https://doi.org/10.1111/j.1749-6632.2011.06439.x

26.

Lenth

Singmann

Love

Buerkner

Herve

(2019). emmeans: Estimated marginal means, aka least-squares means. https://CRAN.R-project.org/package=emmeans

27.

Loimusalo

Huovinen

(2018). Memorizing silently to perform tonal and nontonal notated music: A mixed methods study with pianists. Psychomusicology: Music, Mind, and Brain, 28(4), 222–239. https://doi.org/10.1037/pmu0000227

28.

Loimusalo

Huovinen

Puurtinen

(2019). Successful approaches to mental practice: A case study of four pianists. Music Performance Research, 9, 101–127. http://www.mpronline.net/Issues/Volume%209%20[2019]/MPR%200130%20Loimusalo%20(101-127).pdf

29.

Nakamura

(1987). The communication of dynamics between musicians and listeners through musical performance. Perception & Psychophysics, 41(6), 525–533. https://doi.org/10.1121/1.2026770

30.

Neuhaus

(1993). The art of piano playing. (Originally published 1973). Kahn & Averill.

31.

Palmer

(1989). Mapping musical thought to musical performance. Journal of Experimental Psychology: Human Perception and Performance, 15(2), 331–346. https://doi.org/10.1037/0096-1523.15.2.331

32.

Penttinen

Huovinen

Ylitalo

A.-K.

(2013). Silent music reading: Amateur musicians’ visual processing and descriptive skill. Musicae Scientiae, 17(2), 198–216. https://doi.org/10.1177/1029864912474288

33.

R Core Team (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.

34.

Repp

B. H.

(1992). Diversity and commonality in music performance: An analysis of timing microstructure in Schumann’s “Träumerei”. The Journal of the Acoustical Society of America, 92(5), 2546–2568. https://doi.org/10.1121/1.404425

35.

Repp

B. H.

(1995). Expressive timing in Schumann’s “Träumerei”: An analysis of performances by graduate student pianists. The Journal of the Acoustical Society of America, 98(5), 2413–2437. https://doi.org/10.1121/1.413276

36.

Repp

B. H.

(1996). The dynamics of expressive piano performance: Schumann’s “Träumerei” revisited. Journal of Acoustical Society of America, 100(1), 641–650. https://doi.org/10.1121/1.415889

37.

Revelle

(2019). psych: Procedures for psychological, psychometric, and personality research. Evanston, IL: Northwestern University. https://CRAN.R-project.org/package=psych

38.

Van Vugt

F. T.

Jabusch

H.-C.

Altenmüller

(2013). Individuality that is unheard of: Systematic temporal deviations in scale playing leave an inaudible pianistic fingerprint. Frontiers in Psychology, 4, 134, 1–10. https://doi.org/10.3389/fpsyg.2013.00134

39.

Watson

R. A.

(1995). Representational ideas: From Plato to Patricia Churchland. Springer.

40.

Wilson

T. D.

(2009). Know Thyself. Perspectives on Psychological Science, 4(4), 384–389. https://doi.org/10.1111/j.1745-6924.2009.01143.x

41.

Wöllner

(2013). How to quantify individuality in music performance? Studying artistic expression with averaging procedures. Frontiers in Psychology, 4, 361, 1–3. https://doi.org/0.3389/fpsyg.2013.00361

42.

Yin

R. K.

(2018). Case study research and applications: Design and methods (6th ed.). Sage.