What is your thought process during sight-reading? Advanced sight-readers’ strategies across different tonal environments

Abstract

Sight-reading strategies used for reading music in different tonal environments are critical for musicians to perform accurately. We investigated the strategies advanced sight-readers utilize when they encounter different tonal environments. After a brief study period, six advanced sight-readers performed a through-composed piece that included tonal, non-tonal, and ambiguously tonal music. Immediately following the performance, participants went back through the music and described their thought process and strategies. Participants reported different strategies from one another to play accurately, but there were also common strategies used for different tonal environments. We found verbal reports of critical and relevant strategies of advanced sight-readers were categorized as attention, static analysis, informed intuition, and performer’s analysis. The sight-readers executed their performance interacting with static analysis, informed intuition, and performer’s analysis. Most importantly, participants relied heavily on performer’s analysis for the expressive performance in the tonal section, followed by the non-tonal and ambiguously tonal sections. Findings imply that advanced sight-readers’ strategies moved back and forth between attention, intuition, and analytical strategies based on the demands in each tonal environment.

Keywords

sight-reading strategy tonal environments intuition performer’s analysis thought process

Reading music at sight as accurately as possible seems essential for trained performers to actively engage in varied musical experiences and efficiently learn a variety of repertoire. In general, in music institutions of higher education, sight-reading is considered an important aspect of advanced musicianship and of determining a student’s skill level and artistry. While music majors considered sight-reading an essential skill for developing musicianship, they reported poor music reading skills and expressed a lack of confidence in their sight-reading ability, facilitating a greater focus on sight-reading strategies in this study (Kostka, 1997; Zhukov, 2014).

Other studies have explored the cognitive properties experienced performers display while sight-reading. Researchers suggested highly skilled sight-readers have greater eye–hand span compared with less skilled sight-readers (Furneaux & Land, 1999; Goolsby, 1994a, 1994b; Huovinen, Ylitalo, & Puurtinen, 2018; Penttinen, Huovinen, & Ylitalo, 2015; Rosemann, Altenmüller, & Fahle, 2016; Sloboda, 1974). These readers (compared with less proficient readers) looked farther ahead and then back between points where melodic patterns separated, showing relatively active saccades (movements between visual fixations; for a review, see Puurtinen, 2018). Lim et al. (2019) found advanced sight-readers regulated the distance they looked ahead as a strategy compared with less proficient readers. In other studies, proficient readers showed consistent eye movement patterns with more frequent and shorter fixations in specific areas of tonal music than less proficient readers (Waters & Underwood, 1998). That is, the strategy of the most proficient group of sight-readers was to chunk groups of notes rather than to visualize notes one by one. Similarly, findings showed skilled sight-readers were more dependent on utilizing pattern recognition and prediction skills in their tonal music repertoire compared with less skilled sight-readers (Fine, Berry, & Rosner, 2006; Goolsby, 1994a; Sloboda, 1985). Mishra (2014a) suggested that as musicians develop, they may find that “cues in the notation and aural cues from the performance may interact with music knowledge during sight-reading, resulting in sophisticated guessing” (p. 461). These findings indicate that experienced performers focus on recognizing the structure of tonal music while performing at sight. Furthermore, Kopiez and Lee (2006) found that on the most difficult sight-reading tasks, sight-reading expertise level, speed of information processing, and psychomotor speed were all important factors in successful sight-reading.

Researchers have also explored sight-reading skill for pianists and found these musicians’ understanding of musical characteristics, such as tonal structure, resulted in more accurate performances when reading tonal music (Grutzmacher, 1987; Henry, 2004). There is also evidence that accompanying majors were consistently better at sight-reading than piano performance majors (Lehmann & Ericsson, 1993, 1996). Researchers have also noted the inability to execute expressive performances in sight-reading except by the most advanced readers (Ronkainen & Kuusi, 2009).

Only a few studies have examined sight-reading atonal or non-tonal musical passages. Lewandowska and Schmuckler (2019) found significant differences in sight-reading accuracy of tonal and atonal passages. With Bach chorales and 20th-century music (adapted to represent the three types of tonality), participants made more errors in atonal passages compared with tonal. In general, participants spent more time studying atonal passages prior to performance, played at a slower tempo, and made more errors. Other sight-reading studies of non-tonal music explored specifically the interference of pattern expectancy and memory recall (Fine et al., 2006; Halpern & Bower, 1982; Sloboda, 1984) but did not discuss performers’ strategies.

Ronkainen and Kuusi (2009) observed five professional pianists’ sight-reading of a tonal piece and a non-tonal piece and then interviewed the pianists. The questions and analysis concentrated on the relationship of the keyboard to visual, auditory, and kinesthetic modalities. Performers in the tonal section rarely looked at the keyboard and reported that they did not need to look at the keyboard because they had an aural image and imagined or felt the keys in their head. In contrast, participants looked at the keyboard in the non-tonal section since they were not familiar with the tonal structure or function of intervals. In addition, Bogunović and Vujović (2012) examined kinds and levels of problem-solving strategies of 89 undergraduate music majors. Participants revealed several levels of complexity including deliberate decision-making and strategies relying on intuition. Almost half the participants reported forming an overall idea of the melody, relying on learned patterns and recognition of tonal functions, key structure, and temporal organization. Although participants also reported difficulty performing alterations, leaps, and modulations or losing the tonal center/key, their reported strategies mostly related to using knowledge of music theory to find solutions.

Decision-making processes (both intuitive and analytical) in preparation of music for performance or sight-reading have also been examined. Bangert (2009) interviewed an experienced violinist about general decision-making strategies when preparing music for performance. He found decisions were based on intuition, score analysis, previously learned knowledge, technique, and experience. In addition, Bangert, Schubert, and Fabian (2009) observed a violinist’s decision-making strategies during sight-reading, practice, and performance. Similarly, the violinist drew on mostly intuitive judgments during sight-reading, relying on previous knowledge for instant decision-making and execution. During preparation of the same piece for a performance, the violinist used analytical reasoning with conscious awareness to make decisions but still mostly utilized intuition (unconscious awareness). Bangert, Schubert, and Fabian (2014) introduced a “spiral model of musical decision-making” combining previous music studies and ideas from psychological theories and dual-process theories of cognition. This model suggests that the proportion of intuitive to deliberate decisions change with increasing expertise. Musicians utilize mostly intuitive process at the early stages and move toward more deliberate analytical techniques that are applied during practice. As expertise increases, these learned analytical decisions develop into an automatic process, allowing the musician to move back to a more intuitive process.

It is worth noting the definitions of intuition to understand intuitive process in performance. Betsch (2008) defines intuition as a thinking developed generally from acquired knowledge, which is automatically processed, creating a feeling contributing to decision-making. In the context of music performance, Rink (1990) termed this idea informed intuition, recognizing the importance of calling upon previously learned musical knowledge and experience in the performance process. Furthermore, Rink (1990, 2002) focused on the interaction between intuitive and analytical approaches to music performance. He defined this performer’s analysis as utilizing contextual understanding of music and intuition to shape the music expressively at the moment of performing. Although this view was discussed in the context of music performance, it calls for an investigation focusing on the intuitive or analytical approach to sight-reading, which may provide clues for understanding sight-reading strategies of experienced performers.

Previous studies have focused on the role of intuition and analysis in music performance. Researchers have observed the nature of experienced performers’ music practice to identify individual strategies and conducted interview and found experienced musicians used intuitive (i.e., unconscious process to shape music expressively), analytical (i.e., discovering underlying meaning of music based on knowledge), and versatile (i.e., mixed approach to practice using both intuition and analysis) strategies (Hallam, 1995a, 1995b; Miklaszewski, 1989). Yet, this finding was for a performer’s typical music practice, not for sight-reading. In this study, we adopted music performance into the context of sight-reading. Although studies have revealed performers’ reliance of learned knowledge to accurately execute tonal music in sight-reading, little is known about strategies used to sight-read non-tonal and ambiguously tonal music. In addition, only a few studies ask participants to report their own thoughts and strategies to perform accurately on a sight-reading task. The purpose of this study was, therefore, to investigate self-reported strategies of advanced sight-readers in different tonal environments including tonal, non-tonal, and ambiguously tonal music through the lens of music performance strategies.

Method

Procedure

We asked the College of Music at Yonsei University to distribute a recruitment email to 32 professional pianists with accompanying experience. We received responses from 12 pianists with more than 20 years of performance experience and at least 10 years of accompanying experience. These pianists were either lecturers or part of the doctoral program and consented to participate in the study. To confirm sight-reading proficiency before proceeding with the study, we had each participant perform a sight-reading achievement test.

The music selected for this test was mm. 1–31 from “Trolltog” by Edvard Grieg (1843–1907), published in Book 5 of Lyric Pieces Op. 54 No. 3. “Trolltog,” marked Allegro moderato and in D minor, includes frequent deviations from a traditional tonal idiom. Some of these are post-tonal idioms such as parallel fifths, tritone relations between adjacent chords, and major–minor chords. The piece includes three different tonal environments—tonal, non-tonal, and ambiguously tonal—though each occurrence is brief.

We met each participant individually in a classroom. Two piano professors, each with over 25 years of experience, attended the live performance to assess the accuracy of pitches, rhythms, and to evaluate the entire quality of the performance. Before the pianists played, we confirmed that the participants had never performed the music. Each pianist was then given 90 s to look over the score before performing for the jury.

Based on the scores and discussions with the evaluators, we determined that only the top six performers had the skill level to move forward in the study. These participants were females aged 32–45 (M = 38.86) who started piano lessons between 4 and 7 years old (M = 5.57) and had 27–41 years (M = 33.29) of performance experience. In addition, these pianists had accompanied professionally for 10–23 years (M = 13.86).

For the next part of the study, participants sight-read a piece composed by the second corresponding researcher. The first section of music features a fugal style and imitative polyphonic texture in the key of F major and conforms to a large-scale tonal plan with the usual underlying tonic-to-dominant motion. The middle section, in contrast, is marked by the absence of tonality. This section is not based on tertian chords but on combinations of various intervals, making it difficult to recognize any hint of tonality. The final section is tonally ambiguous (midway between F major and D minor) in homophonic style, containing two clef changes in the right hand, multiple grace notes, and cross rhythms between triplets and 16th notes. The tonic-to-dominant axis (D-A) witnessed through the entire section conflicts with a new dominant-to-tonic axis (C-F) in the last portion, calling the tonal center of the section into question (see Figures 1 to 3).

Figure 1.

Tonal Section of the Piano Piece for Sight-Reading (mm. 1–8).

Figure 2.

Non-Tonal Section of the Piano Piece for Sight-Reading (mm. 15–18).

Figure 3.

Ambiguously Tonal Section of the Piano Piece for Sight-Reading (mm. 22–30).

Each participant met two of the researchers and the two original evaluators from the first jury in a classroom. We video-recorded the session using a SONY FDR-AX40 camcorder. Participants were instructed to study the score for 90 s and then play through all three sections of the composition as musically and accurately as possible. Immediately following the performance, we asked the participant to describe the thoughts they had during the preparation time and their thoughts and strategies to perform accurately and musically.

In the context of a sight-reading task, performance should be continued without interference. Therefore, using retrospective verbal protocol analysis (Fonteyn, Kuipers, & Grobe, 1993; Kuusela & Paul, 2000) was most appropriate for identifying cognitive strategies for the sight-reading task. To help aid in memory retrieval immediately following a task, we continually prompted the participant to point to places in the notation as they described their thoughts from the performance. The two piano professors evaluated the participants in the same manner as the first sight-reading task.

Data analysis

First, we transcribed the video and performed script analysis to study participants’ reported strategies based on a verbal protocol analysis method (Fonteyn et al., 1993). We looked for emerging categories and themes in the transcripts and found different reported strategies to read music and different levels of thinking. For example, some comments mentioned noticing surface information. Other comments moved to a deeper level of thinking regarding intuitive or analytical strategies. Through many repeated discussions and consultation of previous literature in music performance, music theory, and empirical studies, we chose five different operators to characterize these reported strategies: attention, static analysis, performer’s analysis, informed intuition, and self-relevance.

We selected the term attention for the descriptions indicating performers saw surface information (e.g., dynamic markings, time signatures, tempo markings). We created the term static analysis, adopting the term static from Cook (1994), to describe the theoretical analysis that occurred before or during performing. The word static was chosen because there is no action on the part of the performer. Comments coded as static analysis indicated a participant’s attention to a feature plus a further comment that showed further analysis beyond just seeing the notation (e.g., “From the first measure to the fourth, I viewed it as an I chord”).

We adopted the term performer’s analysis from Rink (1990), who defined this as an action taken based on analytical reasoning to play (or intent to play) with musical expression (e.g., “I attended to the repeated tone and tried to bring that out”). The term informed intuition for statements were described as a feeling, instinct, or prediction (Rink, 1990, 2002). These descriptions were often of a feeling, but sometimes the performer could not quite articulate their thoughts and performed on instinct (e.g., “I predicted this harmonic structure in mm. 8-16, which gave me a much better sense of stability,” or “I felt something developing here and expanding”).

Other comments that were related to oneself were categorized as self-relevance. These comments were not strategies but comments metacognitive or regulatory in nature including self-critique, intentions, general strategies, or new or better strategies (e.g., “Now, I see that I should have gone slower here,” or “I don’t usually read all notation analytically during practice”). See Table 1 for definitions of the operators, more examples of quotations, and referencing literature.

Table 1.

Definition of operators from verbal reports, examples, and supporting literature.

Operator	Operational definitions	Examples	Reference information
Attention	Verbalizations about being attentive to specific factual information in the notation as immediately noticed, such as key signature, time signature, tempo change, section division, and tonality.	I attended to indications such as time signature, tonality, and notable figuration at first. I first noticed the markings showing section division to confirm the entire musical construction. I recognized that the first section consists of four-part texture.	Musicians’ attention was selective in reading music score, moving their visual attention to surface information recognized.
Static analysis	Verbal descriptions of understanding static or structural aspects of music. This operator indicates recognizing textural functions of music from surface observations of the notation at a deeper level with analytical thought.	Here, I saw the first 4 measures as one unit according to the tonal structure. I noticed this figuration was repeated, so I expected that it would be repeated throughout.	Music analysis plays a key role in understanding how music works and occurs at distinct categories including musical contents in notation or music text, and sound image of the performer or composer (Bent & Drabkin, 1987). Rink (1990, 2002) identifies studying textural functions of music in notation different from performer’s analysis that contains both the study of music and informed intuition. Cook (1994) points out music in notation remains static.
Performer’s analysis	Verbal reports include both performer’s understanding of music and a specific comment of their intent or attempt to apply it into shaping music. Performer’s analysis integrates intuitive and analytical approach to music performance.	I tried to express the rhythmic motive as important in the first phrase of the second section so I was not concerned of some missing notes as long as I kept the rhythmic motive in flow.	Performer’s analysis is achieved by integrating performer’s intuition termed as “informed intuition,” developed by performer’s prior music knowledge and experience (Rink, 1990, 2002).
Informed intuition	Verbal reports that includes performer’s feeling, instinct, or prediction during the performance.	I did not analyze the entire music theoretically, but got the harmonic flow intuitively.	Rink (1990, 2002) proposes informed intuition, recognizing the importance of calling upon previously learned musical knowledge and experience in the performance process.
Self-relevance	Verbalizations related to self of a metacognitive or regulatory nature including self-critique, intentions, general strategies, or better strategy.	I should have gone slower than the tempo I selected. I should have stood the inner voice out for better musical flow. I found it difficult to read the second section compared to the last section.	Morin (2017) defines self-relevance as “any information that refers to, is relevant to, the self” (p. 2).

For reliability, the participants reviewed the transcriptions of their reports and confirmed both the accuracy of the transcription and our interpretation of the meaning of the statement (member checking). Then, independently of one another, we coded the transcripts using the operators we had chosen. We counted an agreement when all three researchers agreed on the code and a disagreement when all three did not agree. We discussed the disagreements and came to 100% agreement on the coding. Then, we asked an experienced external researcher to independently code 25% of the verbal reports resulting in an agreement of 92.85%.

Results

We analyzed the verbal data relating to the five operators—attention, static analysis, performer’s analysis, informed intuition, and self-relevance—across the preparation time and the three sections. The comments on each operator from the participants were counted for each section for a total of 172 comments (attention = 37, static analysis = 32, performer’s analysis = 37, informed intuition = 18, self-relevance = 48).

Preparation

Participants described what they thought about during their preparation time (36 comments). When describing preparation strategies, participants made 19 comments for the tonal section, 7 for non-tonal, and 6 for the ambiguously tonal section (see Figure 1).

Attention

Participants made 20 comments categorized as attention. We found that participants focused almost entirely on the basic critical markings pertaining to the entire piece, including the time signature changes, the rhythmic unit changes, tempo markings, and how these elements divided the piece into sections:

As I looked at the key signature changes, I found that there are 3 sections and I thought about this. (Preparation, all sections; P4)

And repeatedly, I checked tempo and tonality changes. (Preparation, all sections; P5)

Static analysis

Comments (n = 5) categorized as static analysis were made about the tonal section only during preparation. The language used by participants (“figure repeating,” “a bit Bach,” “like a fugue,” “counterpoint-like”) indicated the participants’ involvement in analytical thought of how the music worked based on previously learned musical knowledge:

As I noticed the figure repeating, I thought that this may be the most noticeable pattern, which guided me to focus on this pattern. (Preparation, tonal; P1)

I noticed that this is F major with one flat. In the middle, it appeared that there was no key signature and returned to the F major again. (Preparation, tonal; P3)

Self-relevance

Only Participants 1 and 3 made comments we categorized as self-relevance (two comments each). The participants conveyed their evaluation of technical difficulty or explained a general strategy they planned to use. Based on that, we summarized the main ideas of their comments in self-relevance to technical evaluation, performance evaluation, or general strategy:

I did not notice there was a bass clef for my right hand at all [during preparation] until I performed. [laughter] (Preparation, ambiguously tonal; P1)

At first, I thought the first part is a little bit like easy but as I read through, I realized that it gets complex. (Preparation, ambiguously tonal; P3)

Performance

When discussing thoughts during performance, participants made a total of 136 comments (tonal section = 78, non-tonal = 24, ambiguously tonal = 34). Refer to Figure 4 for the distribution of comments by participant per operator in each of the sections (tonal, non-tonal, and ambiguously tonal; see Figure 5).

Figure 4.

Comments for Preparation Time by Participant per Operator in Different Tonal Environments.

Figure 5.

Comments for Performance Time by Participant per Operator in Different Tonal Environments.

Attention

Participants reported only 10 comments categorized as attention about their thoughts during the performance (tonal = 8, non-tonal = 0, ambiguous = 2). In contrast to the preparation time, we found that all but two comments in the tonal section were about markings for expressivity (e.g., “fortissimo,” “ritardando,” “slur,” “articulation”) rather than the structural information of the music score. No comments categorized as attention were made in the non-tonal section. The two comments made in the ambiguously tonal section were about structural or basic information, not expressive markings:

I focused on the fortissimo as soon as I turned the page over. (Performance, tonal; P3)

Yes, when I look at the score here, my vision followed the articulation (Performance, tonal; P5)

I looked at the right hand of the score. (Performance, ambiguously tonal; P1)

This is where I saw the adagio marking, and I thought of something different from the previous part. (Performance, ambiguously tonal; P2)

Static analysis

Participants made 27 comments categorized as static analysis about their thoughts during the performance (tonal = 11, non-tonal = 5, ambiguous = 11). In the tonal section, the comments were about the structural elements and, more specifically, some comments were about the fugal elements and tonal relationships. Of all the 27 comments, only 2 comments were about expressive elements; these were made by Participant 3. All other comments were related to tonal, melodic, and rhythmic elements:

I witnessed a textural contrast between sections. Especially in the first section, I saw a fugal procedure and tonic-to-dominant flow, which seems a typical tonal scheme of the fugue and invention. (Performance, tonal; P6)

I focused my vision moving along the subject because like in Bach the subject is going on in the right hand and it is linked to the left hand. (Performance, tonal; P5)

Here, at the beginning, this tone, two eighth notes, sixteenth notes were alternatively repeated. (Performance, non-tonal; P4)

Part B here was like a modern piece because there are many leaps. (Performance, non-tonal; P4)

The first thing I noticed was the characteristic embellishments. Embellishments and dotted rhythms were something here I focused. (Performance, ambiguously tonal; P3)

At this adagio, I thought, is there a theme? I was thinking of that. And then I looked at the beginning part again. (Performance, ambiguously tonal; P6)

Performer’s analysis

Participants made 37 comments categorized as performer’s analysis about thoughts that occurred during the performance (tonal = 24, non-tonal = 10, ambiguous = 3). There were two types of performance analysis statements. The first type included examples where participants pointed out and analyzed rhythmic, melodic, structural, or tonal elements followed by a statement about performing an expressive action not marked in the score (“slow down,” “bring out,” “change articulation,” and so on; tonal = 18, non-tonal = 9, ambiguously tonal = 2). The second type included examples where participants pointed out an expressive marking in the score and stated that they performed that marking (tonal = 6, non-tonal = 1, ambiguously tonal = 1). In addition, some participants gave a detailed explanation about the purpose of the marking or discussed making an extra effort to follow the marking carefully:

Here is the ritardando, so I was thinking about how I shape or manipulate the tempo smoothly. (Performance, tonal; P1)

Obviously, I saw D minor chord, and I thought that I should contrast the tone from the first F chord. (Performance, tonal; P4)

I expressed a little forte here naturally when the notes were wider. (Performance, non-tonal; P4)

I focused on legato but added a little staccato. (Performance, non-tonal; P5)

Here, this kind of embellishment and dotted rhythm, these things were again something characteristic. I performed, making that noticeable. (Performance, ambiguously tonal; P2)

I wanted to perform that ornament very clearly and musically. (Performance, ambiguously tonal; P4)

Informed intuition

Participants made 18 comments categorized as informed intuition about thoughts during the performance (tonal = 11, non-tonal = 2, ambiguous = 5). We categorized comments as informed intuition when participants described a feeling to perform a certain way. Often, these comments did not have a clear explanation and included the words “intuitively,” “felt,” and/or “predict.” Comments in the tonal section revolved around the tonality and the mood or atmosphere of the composition. In particular, participants described a sense of tonality across sections, although they only commented on tonal center in the tonal and ambiguously tonal sections. Also, participants provided descriptions of voices, whether it be expanded or reduced, in the tonal and ambiguously tonal sections only, whereas in the non-tonal section, they reported the absence of tonality or that the mood had changed from the first section:

This is one of the intuitive aspects of this, I attend to down beat rather than for each one to be important. (Performance, tonal; P3)

But I intuitively knew a transition coming little bit later. I though intuitively that there is something a little climax or something like this. (Performance, tonal; P4)

I didn’t feel a tonality in the beginning of meno mosso section. (Performance, non-tonal; P1)

I felt the mood had changed from the Bach [style]. (Performance, non-tonal; P1)

I felt that voices were reduced as meter changed. (Performance, ambiguously tonal; P2)

I felt at the end a feeling of F major but it was not stable. (Performance, ambiguously tonal; P6)

Self-relevance

Participants made 44 comments categorized as self-relevance about the thoughts they had during the performance (tonal = 24, non-tonal = 7, ambiguous = 13). We categorized comments as self-relevance when participants described a specific approach or strategy, reported specific challenges, critiqued how they played, reported what they should have done to perform better, or made comments about the piece or performance:

The part of the octave was uncomfortable because I couldn’t reach my hand over octave, so the rhythm broke down, missing notes. (Performance, tonal-technical evaluation; P1)

I should’ve brought out the inner voice here. (Performance, tonal-performance evaluation; P6)

It is usually possible to play a little smoother when I grasp the regular pattern. (Performance, non-tonal-general strategy; P4)

It is important to keep the rhythm and get the correct sound after all and not to be distracted (from other musical factors) and keep the tempo consistent. (Non-tonal—general strategy; P3)

It is important to keep my tempo consistent. (Ambiguously tonal-general strategy; P6)

Oh, I knew that here, there is F clef, but I do not think it was a turnaround at first. It was a bit difficult. (Ambiguously tonal-technical evaluation; P4)

Discussion

The purpose of this study was to investigate sight-reading strategies of advanced pianists playing different tonal environments. We asked participants to sight-read a newly composed piece with three sections—tonal, non-tonal, and ambiguously tonal—and then report their thought process immediately following the task. The reported thought process of these participants provided substantial clues for how advanced pianists strategically approach reading music within different tonal environments, where certain patterns may or may not be recognizable.

Consistent with findings in other sight-reading studies (such as Mishra, 2014a, 2014b), we found that participants used different strategies from one another to play accurately. For example, Participant 6 was the only participant who mentioned an overall focus on playing the notes and rhythms accurately. Participants 4 and 5 were most similar in their strategies, both discussing the fugal elements, but Participant 4 focused on being expressive and Participant 5 focused on written articulations within the overall structure of the piece. Participant 3 worked to keep the beat very consistent and when she found a rhythmic pattern or melodic pattern, she tried to bring those out in the piece. Participants commented that the unusual structure made this piece very difficult. Although they focused on different specific elements, there were common strategies used within each individual focus applicable to the different tonal environments. We found the strategies of attention, static analysis, informed intuition, and performer’s analysis to be common and critical strategies for these advanced sight-readers. The comments categorized as attention showed that participants noticed critical information quickly and skipped other information, possibly showing the performer filtered out information requiring less cognitive strategy to focus more on difficult passages. In fact, a few participants reported performing without any thoughts in some parts because not all notes were technically and musically demanding. Participants may have overlooked well-understood information because little thought process was required to apply the visual information to the performance. These findings suggest a possibility that participants only verbalized their strategies in places where they focused on technically and cognitively difficult sections. This observation seems to align with research showing experts have shorter and fewer fixations on task-redundant areas in both music and other performance tasks (Drai-Zerbib, Baccino, & Bigand, 2012; Gegenfurtner, Lehtinen, & Säljö, 2011; Goolsby, 1994a).

Comments categorized as static analysis varied in content across the participants and sections. In the tonal section, most comments were about structural aspects of music, such as fugal style related to key and texture. In the non-tonal and ambiguously tonal sections, participants commented on the tonality and non-harmonic tones and noticed repeated rhythmic elements, leaps typical feature of modern music, and wider registers and embellishments, focusing on more micro-levels of tone relation.

As previous studies have discussed in terms of experienced musicians’ reliance on musical knowledge, the verbal reports of static analysis in this study showed participants efficiently matched their prior knowledge to recognize events in the music (Fine et al., 2006; Sloboda, 1976). Since all participants were familiar with tonality and stylistic elements as common structural elements, they may have searched for these relationships first and then when unsure (e.g., in the non-tonal and ambiguously tonal sections) found different ways to perform accurately. For example, when the participants recognized the fugal style, they made comments about using this knowledge to play the section. In fact, Participant 3 stated, “In general, if there are some patterns like in this tonal section, I can perform fluently with few mistakes.” Participant 6 stated, “I feel like it is hard when the tonality is not clear or ambiguous, because I cannot predict what is coming in the next passage. That was the most difficult part.” The different tonal environments influenced participants’ attempts to recognize structural aspects of music when tonality was clear, but when tonality was unclear, their focus was on the detailed component at micro-levels. Similarly, Fournier, Moreno Sala, Dubé, and O’Neill (2019) discussed that simple relationships may be seen at a glance, but when the music no longer makes sense, sight-singers would revert back to more simple decoding strategies.

In all the statements categorized as informed intuition, participants discussed feeling a mood, texture, or tonality change. Similarly, Bangert, Fabian, et al. (2014) determined intuitive music decision-making with performers’ reports, which included words such as “feeling” or “sense,” and Rink (2002) emphasized the performer’s knowledge and experience in recognizing informed intuition during performance. In line with that, we found performers relied on their knowledge on the tonal structure and some participants utilized predictive skills (e.g., harmonic development) as they had the tonal structure or harmonic flow in their mind. Furthermore, some comments described how that intuition was used to shape the music further or play accurately (e.g., “I did not analyze the entire music theoretically, but got the harmonic flow intuitively” and “This is one of the intuitive aspects of this, I attend to down beat rather than for each one to be important”). This was consistent with findings that experienced sight-readers utilize prediction skills in performing a tonal music repertoire (Fine et al., 2006; Goolsby, 1994a; Sloboda, 1985). Although participants’ sensibility of tonal center or change was also applied in the non-tonal or ambiguously tonal section, the comments were limited to how the participants felt the tonal center. This provides a possibility that recognition of tonal center gives performers a sense of stability, which makes prediction possible; otherwise, such a strategy may not be viable. The effect of feeling tonal center during sight-reading thus requires future investigation beyond the interplay between prediction skills and informed intuition.

We also noticed two different levels of performer’s analysis not discussed in previous literature of music performance or sight-reading. Some comments quite simply stated noticing an expressive marking and performing that marking (n = 9). The majority of comments were about adding expressions not explicitly marked in the score to “bring out” a certain musical element that had been analyzed. In addition, in the tonal section, 18 of the 24 comments categorized as performer’s analysis were observations made at this deeper level. In the non-tonal section, the content of 8 of 10 comments was at this deeper level, and in the ambiguously tonal section, 2 of the 3 were deeper level comments on performer’s analysis (e.g., analysis on rhythmic, melodic, structural, or tonal elements followed an expressive action not marked in the score) rather than just statements that the participants performed what was marked on the page. The majority of participants made comments at both levels. This provides evidence that advanced sight-readers engage in analytical strategies needed to perform different tonal environments, communicating with their intuition for expressive performance. This contrasts the findings of other researchers, such as Bangert et al. (2009), who found that only intuition was used during sight-reading. Thus, the conception of performer’s analysis may play a key role in discovering advanced sight-readers’ analytical strategies. In addition, the presence of performer’s analysis at a deeper level in sight-reading gives rise to the notion that performer’s analysis could be expanded according to the types of music performance (Rink, 1990, 2002).

Following a 90-s period of silent preparation, participants played the piece in full. Most of the strategies were reported in the preparation time for the tonal section. Furthermore, preparation time may have played a key role in utilizing performer’s analysis for sight-reading. For example, the advanced sight-readers recognized structural aspects of the music score during preparation which may have influenced the ability to shape expression during performance. Few comments were made about the non-tonal and ambiguously tonal sections for preparation time, and those comments were only about the visible markings (attention).

Although all participants were advanced sight-readers, they did not all perform at the same level of accuracy. Participant 6 was the most expressive performer and was the strongest sight-reader on this task in terms of accuracy. She also made the fewest comments compared with others but discussed her focus on accuracy of the notes and the written markings. Maybe Participant 6 was so advanced that adding expressive elements had become more automatized as discussed in the spiral model theory (Bangert et al., 2014). Further research is warranted to study the relationships between performance quality and the cognitive strategies used to play accurately in expert sight-readers using a validated assessment tool to measure performance quality. We hypothesize that the participants with the highest scores of these expert sight-readers are more likely to rely on intuition to play expressively and accurately.

Participants may have spent more time looking at the tonal section since it was the first section in the piece, possibly resulting in more descriptions in the tonal section compared with non-tonal or ambiguously tonal sections for both preparation and performance. It is also possible that 90 s was not enough time for all participants to look past the tonal section. Only one participant indicated they did not look over the entire piece, and no one requested more time. Furthermore, all but one participant made at least one comment during the recall of preparation strategies for the ambiguously tonal section. Future research could have participants speak out loud during preparation time or report strategies immediately after silent study before performing. Furthermore, preparation time could be added between sections, and the sections could be presented in a random order.

We collected retrospective verbal reports that may have resulted in memory lapses due to the time gap between the performance and reported strategies. To counteract this issue, we provided the music and asked each participant to point to the places in the music and describe only what they had thought about as they were performing. In future studies of verbal protocol analysis, participants could play a shorter piece of music or stop between sections to report thought processes instead of completing the entire through-composed piece in one performance.

Fournier et al. (2019) in their inventory for cognitive strategies in sight-reading discussed that in their own experiences, improvisation was a common strategy used to sight-read and to teach sight-reading. However, they did not find any descriptions of use of improvisation in their review of textbooks, studies, and interview transcripts. Similarly, no participants in the current study discussed improvising certain parts of the music as they focused on another part. Further research could specifically explore the use of improvisation as a strategy for sight-reading and as a teaching method to improve sight-reading skill.

The results of this study have implications for music educators as well as professional performers. In this study, some participants reported what they should have done to perform better, showing self-regulation and cognition for improvement in future tasks. McPherson (1994) found a similar self-regulatory approach through comments from adolescent brass and wind players about how the participants prepared a sight-reading task. In fact, one of the better sight-readers in that study mentioned being instructed to always say aloud the key signature and time signature before playing. It is possible, speaking aloud key features and strategies may improve self-regulation strategies, therefore improving sight-reading ability. Therefore, teachers could utilize the method in the current study and instruct students to describe their focus and thought processes before and after sight-reading, guiding them to recognize strategies like attention, static analysis, performer’s analysis, and intuition that best allow each individual to sight-read more accurately and expressively. Through these reflections, music students and professional performers may be able to acquire sight-reading skills found in more advanced sight-readers. Further research testing the effect of speak aloud strategies on sight-reading performance is warranted.

The thought processes of these participants provided substantial clues for how advanced pianists strategically approach reading difficult music within different tonal environments. We found the strategies of attention, static analysis, informed intuition, and performer’s analysis as common and critical strategies in these advanced sight-readers. The pianists’ comments clearly showed their thought processes moved back and forth between attention, intuition, and analytical strategies based on the demands in the musical score. These findings contribute a more specific understanding of cognitive strategies utilized by advanced sight-readers to perform expressively and accurately.

Footnotes

Acknowledgements

We would like to thank the pianists at Yonsei University for participating in this experiment. We also wish to thank the reliability reviewer for their work on categorizing the comments of this study.

Funding

The author(s) disclosed receipt of the following support for the research, authorship, and/or publication of this article: This study was supported by the Yonsei Research Grant of 2020.

ORCID iDs

You Jin Kim

Rebecca Atkins

References

Bangert

(2009). Doing without thinking? Aspects of musical decision-making. In Williamon

Pretty

Buck

(Eds.), Proceedings of the International Symposium on Performance Science (pp. 403-408). Utrecht, The Netherlands: European Association of Conservatoires.

Bangert

Fabian

Schubert

Yeadon

(2014). Performing solo Bach: A case study of musical decision-making. Musicae Scientiae, 18, 35–52. https://doi.org/10.1177/1029864913509812

Bangert

Schubert

Fabian

(2009). Decision-making in unpracticed and practiced performances of Baroque violin music. In Stevens

Schubert

Kruithof

Buckley

Fazio

(Eds.), Proceedings of Research, the International Conference on Music Communication Science (pp. 3-6). Sydney, New South Wales, Australia: Western Sydney University.

Bangert

Schubert

Fabian

(2014). A spiral model of musical decision-making. Frontiers in Psychology, 5, 1–11. https://doi.org/10.3389/fpsyg.2014.00320

Bent

Drabkin

(1987). Analysis. New York, NY: W.W. Norton.

Betsch

(2008). The nature of intuition and its neglect in research on judgment and decision-making. In Plessner

Betsch

(Eds.), Intuition in judgment and decision-making (pp. 3-22). New York, NY: Taylor & Francis.

Bogunović

Vujović

(2012). Metacognitive strategies in learning sight-singing. Psiholoska Istrazivanja, 15, 115–133. https://doi.org/10.5937/PsIstra1202115B

Cook

(1994). A guide to musical analysis. Oxford, UK: Oxford University Press.

Drai-Zerbib

Baccino

Bigand

(2012). Sight-reading expertise: Cross-modality integration investigated using eye tracking. Psychology of Music, 40, 216–235. https://doi.org/10.1177/0305735610394710

10.

Fine

Berry

Rosner

(2006). The effect of pattern recognition and tonal predictability on sight-singing ability. Psychology of Music, 34, 431–447. https://doi.org/10.1177/0305735606067152

11.

Fonteyn

M. E.

Kuipers

Grobe

S. J.

(1993). A description of think aloud method and protocol analysis. Qualitative Health Research, 3, 430–441. https://doi.org/10.1177/104973239300300403

12.

Fournier

Moreno Sala

M. T.

Dubé

O’Neill

(2019). Cognitive strategies in sight-singing: The development of an inventory for aural skills pedagogy. Psychology of Music, 47, 270-283. https://doi.org/10.1177/030573561774514

13.

Furneaux

Land

M. F.

(1999). The effects of skill on the eye–hand span during musical sight–reading. Proceedings of the Royal Society of London. Series B: Biological Sciences, 266, 2435–2440. https://doi.org/10.1098/rspb.1999.0943

14.

Gegenfurtner

Lehtinen

Säljö

(2011). Expertise differences in the comprehension of visualizations: A meta-analysis of eye-tracking research in professional domains. Educational Psychology Review, 23, 523–552. https://doi.org/10.1007/s10648-011-9174-7

15.

Goolsby

T. W.

(1994a). Eye movement in music reading: Effects of reading ability, notational complexity, and encounters. Music Perception, 12, 77–96. https://doi.org/10.2307/40285756

16.

Goolsby

T. W.

(1994b). Profiles of processing: Eye movements during sightreading. Music Perception, 12, 97–123. https://doi.org/10.2307/40285757

17.

Grutzmacher

P. A.

(1987). The effect of tonal pattern training on the aural perception, reading recognition, and melodic sight-reading achievement of first-year instrumental music students. Journal of Research in Music Education, 35, 171–181. https://doi.org/10.2307/3344959

18.

Hallam

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

19.

Hallam

(1995b). Professional musicians’ orientations to practice: Implications for teaching. British Journal of Music Education, 12, 3–19. https://doi.org/10.1017/s0265051700002357

20.

Halpern

A. R.

Bower

G. H.

(1982). Musical expertise and melodic structure in memory for musical notation. The American Journal of Psychology, 95, 31–50. https://doi.org/10.2307/1422658

21.

Henry

M. L.

(2004). The use of targeted pitch skills for sight-singing instruction in the choral rehearsal. Journal of Research in Music Education, 52, 206–217. https://doi.org/10.2307/3345855

22.

Huovinen

Ylitalo

A.-K.

Puurtinen

(2018). Early attraction in temporally controlled sight reading of music. Journal of Eye Movement Research, 11(2), 1–30. https://doi.org/10.16910/jemr.11.2.3

23.

Kopiez

Lee

J. I.

(2006). Towards a dynamic model of skills involved in sight reading music. Music Education Research, 8(1), 97–120. https://doi.org/10.1080/14613800600570785

24.

Kostka

M. J.

(1997). Effects of self-assessment and successive approximations on “knowing” and “valuing” selected keyboard skills. Journal of Research in Music Education, 45, 273–281. https://doi.org/10.2307/3345586

25.

Kuusela

Paul

(2000). A comparison of concurrent and retrospective verbal protocol analysis. The American Journal of Psychology, 113, 387–404. https://doi.org/10.2307/1423365

26.

Lehmann

A. C.

Ericsson

K. A.

(1993). Sight-reading ability of expert pianists in the context of piano accompanying. Psychomusicology, 12, 182–195. https://doi.org/10.1037/h0094108

27.

Lehmann

A. C.

Ericsson

K. A.

(1996). Performance without preparation: Structure and acquisition of expert sight-reading and accompanying performance. Psychomusicology, 15, 1–29. https://doi.org/10.1037/h0094082

28.

Lewandowska

O. P.

Schmuckler

M. A.

(2019). Tonal and textural influences on musical sight-reading. Psychological Research. Advance online publication. https://doi.org/10.1007/s00426-019-01187-1

29.

Lim

Park

J. M.

Rhyu

S-.

Chung

C. K.

Kim

S. W.

(2019). Eye-hand span is not an indicator of but a strategy for proficient sight-reading in piano performance. Scientific Reports, 9, Article e17906. https://doi.org/10.1038/s41598-019-54364-y

30.

McPherson

G. E.

(1994). Factors and abilities influencing sightreading skill in music. Journal of Research in Music Education, 42, 217–231. https://doi.org/10.2307/3345701

31.

Miklaszewski

(1989). A case study of a pianist preparing a musical performance. Psychology of Music, 17, 95–109. https://doi.org/10.1177/0305735689172001

32.

Mishra

(2014a). Factors related to sight-reading accuracy a meta-analysis. Journal of Research in Music Education, 61, 452–465. https://doi.org/10.1177/0022429413508585

33.

Mishra

(2014b). Improving sightreading accuracy: A meta-analysis. Psychology of Music, 42, 131–156. https://doi.org/10.1177/0305735612463770

34.

Morin

(2017). Toward a glossary of self-related terms. Frontiers in Psychology, 8. https://doi.org/10.3389/fpsyg.2017.00280

35.

Penttinen

Huovinen

Ylitalo

A.-K.

(2015). Reading ahead: Adult music students’ eye movements in temporally controlled performances of a children’s song. International Journal of Music Education, 33, 36–50. https://doi.org/10.1177/0255761413515813

36.

Puurtinen

(2018). Eye on music reading: A methodological review of studies from 1994 to 2017. Journal of Eye Movement Research, 11(2), 1–16. https://doi.org/10.16910/jemr.11.2.2

37.

Rink

(1990). Review of musical structure and performance by Wallace Berry. Music Analysis, 9, 319–339.

38.

Rink

(2002). Analysis and (or?) performance. In Rink

(Ed.), Musical performance: A guide to understanding (pp. 35–58). Cambridge, UK: Cambridge University Press. https://doi.org/10.1017/cbo9780511811739.004

39.

Ronkainen

Kuusi

(2009). The keyboard as a part of visual, auditory, and kinesthetic processing in sight-reading at the piano. In Louhivuori

Eerola

Saarikallio

Himberg

Eerola

(Eds.), Proceedings of the 7th Triennial Conference of European Society for the Cognitive Sciences of Music (pp. 453-458). Jyväskylä, Finland: University of Jyväskylä.

40.

Rosemann

Altenmüller

Fahle

(2016). The art of sight-reading: Influence of practice, playing tempo, complexity and cognitive skills on the eye–hand span in pianists. Psychology of Music, 44, 658–673. https://doi.org/10.1177/0305735615585398

41.

Sloboda

J. A.

(1974). The eye-hand span-an approach to the study of sight reading. Psychology of Music, 2, 4–10. https://doi.org/10.1177/030573567422001

42.

Sloboda

J. A.

(1976). The effect of item position on the likelihood of identification by inference in prose reading and music reading. Canadian Journal of Psychology/Revue Canadienne de Psychologie, 30, 228–237. https://doi.org/10.1037/h0082064

43.

Sloboda

J. A.

(1984). Experimental studies of music reading: A review. Music Perception, 2, 222–236. https://doi.org/10.2307/40285292

44.

Sloboda

J. A.

(1985). The musical mind: The cognitive psychology of music. Oxford, UK: Oxford University Press.

45.

Waters

A. J.

Underwood

(1998). Eye movements in a simple music reading task: A study of expert and novice musicians. Psychology of Music, 26, 46–60. https://doi.org/10.1177/0305735698261005

46.

Zhukov

(2014). Exploring advanced piano students’ approaches to sight-reading. International Journal of Music Education, 32, 487–498. https://doi.org/10.1177/0255761413517038