Competing With Myself at the Piano: Implications for Research and Learning
This paper was presented at the 25th Clifford K. Madsen International Symposium for Research in Music Behavior, Asheville, NC, March 20-23, 2024.
While preparing to move from Cleveland Heights to Albuquerque eighteen months ago, I dispensed with many items. Among them were tape cassettes and reel-to-reel tapes of my solo piano performances. As I hadn’t listened to them in decades and no longer owned a tape recorder, I had no justification for keeping them. I was curious to hear what I had sounded like back then, however, so before I trashed them I had them digitized. The oldest recordings dated from 1973, when I was a music performance MFA student at Sarah Lawrence College. The most recent one came from the last of three solo recitals I gave at McGill University, in 1984.
I was surprised to discover that my long held beliefs about these performances did not match what I was now hearing. The performances I remembered as bad hadn’t been so bad, and the performances I remembered as good weren’t as good as I had hoped they would be. Listening to the recordings thus gave me both relief and disappointment.
One recording interested me more than the others. It was of Scriabin’s Poéme, op. 32 no. 1, and it caught my attention for two reasons: my performance of it was good, and I hadn’t remembered playing it in recital. Certainly I recalled learning the piece, but I had thought that my only performance of it in front of others was at a 1975 audition. Yet there it was on the recording of a 1976 Syracuse University recital; I had played it as an encore.
As I listened to the Poéme a second time, I wondered if I still might be able to play that well. Krampe & Ericsson (1996) found that the average performance level of older expert pianists was only slightly lower than it was for younger expert pianists. They concluded that older pianists were able to maintain their skills because they engaged in deliberate practice throughout their lives. Given that I had kept up my practicing across the years, perhaps I too would be able to play as well as I once did.
My fingers no longer look like they used to. Six are slightly bent at their distal phalanges, and x-rays revealed that all ten are missing cartilage. I have been unable to make a tight fist in either hand since 2020 due to arthritis, and my hand span is smaller than it was when I was younger. In addition, my brain at age 77 undoubtedly looks and works differently than it did when I was 30. I haven’t noticed major negative effects on my piano playing arising from these changes, but perhaps any deterioration has been gradual or minimal enough to avoid detection. Or maybe my technique really hasn’t changed. I could have tested myself more stringently by relearning one of the more technically demanding pieces I had played long ago, say the Liszt Sonata in B Minor, but such an undertaking would have been very time-intensive. Instead, I decided to work on the three-minute Poéme. I was motivated for three reasons: first, I love the music; second, the challenge of trying to match my 1976 performance intrigued me; and third, I would use the opportunity to try some new (for me) ways of working at the piano. This last reason is my justification for presenting this case study, as I believe that what I learned from the experience has implications for both research and learning.
How I practiced
I approach a composition learned years earlier as if I had never played it. I start by practicing interlocking measures: after working on the first two measures until I can play them smoothly, I work on measures two and three, then measures three and four, and so on. I gradually increase the size of practice units to coincide with phrases, sections, and, finally, the full score. I memorize the music as quickly as I can, and I practice predominantly slowly at first, more frequently at final performance tempo later on. I also adhere to some of the ideas presented by ChatGPT in Appendices A and B, such as practicing regularly, practicing hands separately, and recording myself on occasion.[1] In summary, my way of learning and relearning piano music might be considered traditional. I suppose it also fits the definition of deliberate practice, regardless of whether the opposite of deliberate is either mindless or accidental.
My modus operandi changed in 2020. It was then that Covid-19 transformed tens of thousands of musicians into exhibitionists. Like them, I decided that what the world needed were my videoed recordings. Despite the existence of dozens of excellent professional performances on YouTube for any piece I might choose to work on, I was seduced by the possibility that posting my own performances would make a contribution.
An added appeal of my project was that I could play for others without having to experience stage fright. Though the recording process can be nerve wracking, the prospect of unlimited takes makes it bearable. I bought a Zoom Q8 “Handy Video Recorder” and a microphone stand, which along with my Steinway grand piano and living room was all I needed. I felt confident that at a minimum, my playing would improve as I worked toward a polished performance. At a maximum, I would produce recordings good enough to give others pleasure.
I changed my old practice regimen in three ways. The first followed from my purchase of the Zoom 8: I recorded myself incessantly. The second was that I collected data in order to quantify changes in my performance. By timing myself and rating my performances’ technical and musical qualities, I would strive for a series of “personal bests.” In this way I treated piano practice as much as an athletic endeavor as an artistic one.
Neither of these approaches is new. Musicians have been studying their own performances via recordings ever since home sound recording became feasible. Certainly there has been much research using both audio and video to improve almost every aspect of music learning. Both recording oneself and taking data on one’s performance can be considered examples of self-regulation in music learning (Varela et al., 2014), about which there also is much research. However, I couldn’t find any research that employed self-recording and self-data collection as monomaniacally as I did in this study.
The third and final way in which I modified my piano practice was by listening to and observing audio and video recordings of expert performances of the Poéme while I was relearning the work. This activity is controversial within the classical piano community (Piano World Forum, 2011). Some respondents feel that listening to others stunts a student’s creative development. Others insist that the activity could help one make informed decisions about interpretation, especially if the musician is exposed to many different performances rather than one. I used observation both to seek out new interpretations and to learn in what ways my own performances didn’t measure up to professional recordings.
Scriabin’s Poéme op. 32 no. 1
Faubion Bowers, author of a definitive biography of Scriabin (Bowers, 1996), describes the Poéme as follows:
In it [Scriabin] flung magical distributions of tones into that unknown nowhere he so loved, and whence he evoked such startlingly felicitous sonorities. Pleasing quietude caresses the ear here. The beginning is marked con affetto, or with affection. Sabaneeff calls the Poéme “an erotic kiss…a kind of sexual dissolving in waves of emotion.” Magic and affection are present, certainly, but where? Is it in the symmetry?
Unknown nowheres notwithstanding, one might conclude from this description that the Poéme does not require great speed or power. This is true. Instead it presents difficulties in the realms of touch, projection, dynamic control, and rhythm. The Poéme demands the ability to play cross-rhythms such as three-against-five accurately and smoothly, and dynamics must be controlled so that the melody projects while subordinate lines do not become submerged. It’s a contrapuntal work that combines passion with delicacy, and it’s not easy to pull off.
Method
Video-recording
After rememorizing the Poéme, which took about a month, I began video-recording myself. I had recorded infrequently in the past, but for the Poéme I recorded takes on about half of my practice sessions. I’d play through the piece once or twice slowly, work intensively on sections that were not secure, then record myself while playing the piece in its entirety at its final tempo from one to three times. Altogether, I recorded the Poéme about 300 times over eight months.
Record keeping
Quantifying improvement, like a runner or a swimmer keeping track of trial times, or a baseball pitcher using a speed gun to measure pitch velocity, became an important aspect of my practice. I logged timings of completed performances, numbers of mistakes made, and numerical ratings of expressivity. I didn’t statistically analyze these data because I was wary of reliability for the second and third measurements; as for the first, it was obvious from looking at the data that gradually I played the Poéme faster as I continued working on it.
Listening to other performances
I listened to other performances of the Poéme every third or fourth day. One of my most frequently used techniques was to hear in sequence a YouTube performance by a concert pianist, one of my own recent recordings, and a performance by a second concert pianist. The procedure often depressed me, but it did show me clearly what I needed to do to make my version more professional.
Results
Tempo
My first recordings were disappointing not just because I was still making mistakes, but because I was playing too slowly. I was surprised at my slowness. My tempos seemed right when I was playing, and it was only when I listened afterward that I realized how sluggish they were. This divergence between how I perceived tempo while playing versus listening persisted for months.
“Too slow” might be considered a matter of opinion. Certainly there are numerous examples on CD of performances by great pianists that are aberrantly slow. In addition, it is known that performance tempo may vary considerably across pianists (Wapnick, 1987), and even across different performances by one pianist. I had thought that my initial 2023 performances dragged, but were they slow compared to how the Poéme is played by experts? In order to find out, I compiled a list of recordings as performed by 15 select professional pianists. These were chosen from over 100 different recordings of the Poéme I found on Classical Music Online (https://classical-music-online.net) and on YouTube. I added my 1976 recital performance and two video-recordings from 2023 to the list, shown in Table 1[2]. The duration of each recording was established after locating its beginning and conclusion, as recordings sometimes started or ended with dead space or applause.
Table 1: Durations of the Scriabin Poéme on Selected Recordings
Performer Duration (minutes:seconds)
Maria Lettberg 2:22
Christopher O’Reilly 2:22
Daniil Trifonov 2:37
Yunchan Lin 2:50
Heinrich Neuhaus 2:52, 3:08
Joel Wapnick (2023) 2:52, 3:23
Joel Wapnick (1976) 2:55
Valentina Lisitsa 2:57
Vladimir Sofronitsky 3:00, 3:14
Vladimir Horowitz 3:04, 3:16
Yuja Wang 3:10, 3:48
Mikhail Pletnev 3:10
Walter Gieseking 3:12
Vladimir Ashkenazy 3:15
Josef Hofmann 3:20
Joseph Villa 3:20
Anna Fedorova 3:36
Ivo Pogorelich 4:18
With the exception of one of Yuja Wang’s two performances and Ivo Pogorelich’s outlier, my early 2023 recordings (not included in Table 1), which ranged from 3:40 to 3:50, were slower than those of the professional pianists. Indeed, I had been playing too slowly.
My 1976 and 2023 Performances of the Scriabin Poéme
I’ve posted two files of my Poéme performances on YouTube; the links to them are given below.
The 1976 performance is here:[3]
https://www.youtube.com/watch?v=83AI_jFmmFU
If you get an “Advanced Threat Protection” warning, just click the “Skip Verification” button.
My two best (in my opinion) 2023 performances are here:
https://www.youtube.com/watch?v=p2nmrKAkrMk
I’ve included them both because I couldn’t decide which I liked better. They differ considerably in tempo.
The 1976 performance, to my ears, sounds more technically assured than either 2023 performance. Dotted rhythms are played more precisely, and the differences between staccato and legato passages are more pronounced. These advantages are particularly noticeable in the Poéme’s short sixteenth-note passages, which occur eight times throughout the piece.
The sixteenth-notes are also played more softly in the 1976 recording than in the 2023 recordings.
Another impression I received is that in the 1976 recording, differences between loud and soft playing are more pronounced than in either 2023 recording. In order to verify my impression, I would have had to somehow measure loudness variability in each performance, while keeping overall loudness constant. Given the limited technology I had at my disposal, I wasn’t able to do this.
It is interesting that although the duration of the 1976 performance is three seconds longer than it is for the faster of the 2023 performances, the tempos of the first six measures of the Poéme’s two B sections (the first three lines of pages 2 and 4 of the score, respectively) are considerably slower (at 22 and 23 seconds, respectively) than they are in the 1976 performance (19 and 19 seconds, respectively). Tempo variability apparently was greater in the earlier recording.
The second 2023 performance on the video is about 31 seconds, or 17%, slower than the first 2023 performance. It sounds more relaxed than both the 1976 performance and the faster 2023 performance, and at certain times of the day I prefer it over the others. So much for listener reliability.
Discussion
Self-Recording
Though there is a considerable amount of research on audio and video self-feedback, and on self-regulation in music performance, I was unable to find studies that employed self-recording to the degree that I did in this study. I recorded myself regularly for eight months, made hundreds of recordings, and spent much time listening to them. I continue to use self-recording as part of my regular practice because I find it valuable.
Self-recording makes me self-conscious. I spend much more time evaluating my playing when I’ve recorded myself than when I haven’t. In fact I feel as if another person is in the room when the Zoom Q8 is on. Even after many months of recording myself, this feeling has not gone away. During a successful recording, I’ll get progressively more anxious the closer I am to the end of the piece. This of course makes it increasingly difficult to produce error-free recordings. On the other hand, recording oneself may be good preparation for playing in front of an audience.
Record keeping
The data I produced from record keeping was not useful, especially as it may have been unreliable. It is difficult to count mistakes while playing. My overall evaluations (on a 100-point scale) of how well I played after finishing a take may be more reliable than counting errors, but my standards may have changed from month to month. There was an important benefit of record keeping, however: it motivated me. On a day-to-day basis, I have difficulty detecting improvement. Even over longer periods this is not easy—not only because improvement at an instrument comes slowly, but because it is difficult to accurately recall how well I played, say, a month or two earlier.
Listening and watching model performances
Listening and watching recordings and videos by professional musicians made me aware of many details in the music that otherwise I might have overlooked. I adopted a few fingerings from other videos, and made note of the surprisingly frequent moments when professionals played something different from what the score indicated.
Sometimes listening to others made me aware of certain aspects of performance that I wanted to avoid. Many experts injected what I thought was too much rubato into their performances. Others omitted certain details in the music that I felt were indispensable. Musicians can become hypercritical when listening to other musicians play “their” piece, and I am no exception.
Of the three interventions, my guess is that self-recording was most influential in helping me improve. It was also the most cringe-inducing. I hated hearing when I sounded like “two elephants,” as my teacher sometimes described my hands when I played too loudly or clumsily. Even worse was when I sounded too timid. Observation helped me notice mistakes both of omission and commission, but even more important was how listening to my own performances helped me to see if I was making compelling music.
I could have recorded myself without video, but decided to include it primarily to observe how my fingers, hands, arms, and body moved. I don’t think there was any connection between these movements and how well a take sounded. Nor do I believe that there was anything problematic in my technique, as shown in the videos[4].
Performance differences
Was I able to play Scriabin’s Poéme in 2023 as well as I did in 1976? The three performances did not differ much with regard to accuracy. Moreover, the faster of the two 2023 performances was played at the same tempo and with similar rhythmic control as the 1976 performance. Nevertheless, there are a few aspects of the 1976 performance that I didn’t—or couldn’t—emulate 46 years later. Differences between staccato and legatoplaying are more evident in the 1976 recording. Moreover, the earlier recording sounds more improvisatory through its use of a wider dynamic range and greater tempo variability, and by making changes in the music suddenly rather than gradually. Consequently, I believe that the earlier recording achieves a sense of freedom and vividness that is lacking in the recent performances. On the other hand, the slower 2023 performance has a relaxed charm that is perhaps more in line with Faubion Bowers’ description of the Poéme than is the 1976 performance.
I had long known that wide dynamic range in performance contributes to expressivity. Patterson (1974), however, showed that in certain cases a pleasing effect of smoothness could be achieved by playing within a restricted dynamic range[5]. A case might be made that such smoothness is appropriate in the Poéme, given its relatively calm nature. If so, my 2023 performances may be better suited to the musical content of the Poémethan I think they are.
Limitations of the performer
I found it difficult to radically alter my playing after I had learned the Poéme well. I might listen to a take, for example, and resolve to play the following one more quickly. When I would listen afterward, confident that I had succeeded, I would discover that I had shaved off only a second or two from the earlier take. What I thought I had been doing while playing did not match what I actually did. Or I’d be satisfied with a take, thinking that it was at least as good as my 1976 recording. Then I’d listen to the 1976 recording afresh and find that my latest recording had fallen short yet again. My inability to achieve the level of playing evident in the 1976 recording was the main reason I continued practicing a three-minute piece for eight months.
Eventually I was able to match my target speed of 2:55—the duration of my 1976 performance—but it took me much so-called deliberate learning to get there. Perhaps the importance of deliberate learning lies not so much in paying attention as it does in its promotion of a willingness to experiment.
Reliability
A comparison of the artistic value in my Poéme performances must begin with the understanding that, beyond technical proficiency, preferences sometimes change with the wind. I prefer my 1976 performance of the Poéme to either of the 2023 performances most of the time, but not all of the time. I continue to experience some indecision in determining which of my recordings is best. It is possible that even for the most serious of listeners, performance preference may change—not just over months and years, but within a day or even an hour.
I don’t recall how difficult it was for me to learn the Poéme back in the mid-1970s. My guess is that relearning it last year took longer, and was more arduous. Here’s another guess: older pianists learn new pieces more slowly than do younger pianists. They may be more reluctant to learn new pieces as well.
The specific circumstance of the 1976 performance may have resulted in a freer performance than anything I could have produced, either recently or 46 years ago. I had played the Poéme at the end of a long recital, and it was preceded on its second half by two highly emotional and virtuosic pieces: Mussorgsky’s Pictures at an Exhibition and Ravel’s Alborada del gracioso. When it was time to play the Poéme, I may have been unusually inspired and relaxed. Such a potentially idiosyncratic situation shows how my study is subject to the weaknesses of many case studies, in that circumstances may have been unique enough to make the validity of generalizations from it suspect.
Problems of cause and effect
One might consider self-recording, record keeping, and listening and watching model performances to be the three independent variables in this case study. A problem with this formulation is that there is no control group. Indeed, the only real “treatment” here is the passage of time. Musicians change in performance proficiency for many reasons, including physical and cognitive aging, increased knowledge of music, aesthetic development, the influences of other musicians, and changes in the amount of practice. If performances were not to change much over many years, it is unclear whether we should marvel at how resistant they are to the effects of time, or wonder about the musician’s lack of artistic growth.
It would be interesting to gain a greater sense of how musical performance changes with age. There appear to be wide differences between individuals. Most of us have witnessed in concert both musicians who should have quit years ago and musicians who appear to be ageless, for example. Also, knowledge of what aspects of musical technique and expressiveness are most susceptible to the effects of time has not been extensively researched. Knowledge in this respect might be gained by having older accomplished musicians relearn music that they recorded long ago but haven’t practiced recently.
In certain sports (swimming, gymnastics, tennis), peak performance comes early. For musicians, this may or may not be the case. Pianists such as Artur Rubinstein, Vladimir Horowitz, Claudio Arrau, and Menachem Pressler pleased audiences well into their 80s (90s in the case of Pressler). On the other hand, at a purely technical level these pianists were far superior when they were in their 40s. If in fact their later performances can be shown to be more musically satisfying than their earlier ones, it would be interesting to know how the older masters accomplished such miracles. Is superior late-life performance a consequence of having experienced more of life, and of music? Or is this supposition false, and the goodness of a musical performance primarily depends on the performer’s technique?
Implications
The literature on playing the piano is enormous. Hundreds of books and studies have been written on technique, piano teaching, historical performance practice, and practical advice for pianists of all kinds and at all levels. More are filled with preparatory exercises written by the likes of Cramer, Hanon, Czerny, John Thomson, and countless others—to say nothing of technical exercises and etudes—real music!—composed by Bach, Clementi, Chopin, Liszt, and Brahms.
With fecundity comes disagreement, and with disagreement comes the need for answers. Perhaps the most obvious controversy in piano performance relates to technique. Should a pianist play with curved or flat fingers (Zhdanov, 2023)? High or low wrist? How much involvement of the forearms, upper arms, and entire upper body is ideal?
One might question the implied either-or dichotomy inherent in these questions. In sports, there is considerable variability in certain well practiced motions such as tennis serve, baseball pitching and batting, golf swings, and basketball foul shooting. Similarly, it is possible that there is no one right way of playing the piano. Or maybe there is? If we look to experimental research on the topic, we find little to guide us other than motion studies. A partial bibliography of such studies (Turner et. al, 2021), reveals that this research consists primarily of case studies—alike in form though not in the specifics of what I’ve done in this case study.
When it comes to researching how to play the piano, there may be little alternative to the case study. We cannot randomly assign students to treatments in order to answer even basic questions such as whether practicing scales versus appropriate control activities is beneficial, what warmup exercises and technical approaches are most resistant to injury, and whether practicing beyond three, six, or nine hours a day has long-term negative consequences. Indeed, almost all of ChatGPT’s recommendations in Appendix A have come down to us as given truths learned through extensive experience. There is value in them—probably. On the other hand, I haven’t used a metronome in fifty years. What have I been missing?
In conclusion, one case study about piano playing demonstrates nothing. But maybe ten, or a hundred, can.
References
Bowers, Faubion (1996). Scriabin: a Biography. Second edition, Mineola, New York: Dover Publications.
Ericsson, K.A., Krampe, R.T., & Tesch- Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100(3), 363-406.
Hallam, S. (1997). The development of metacognition in musicians: Implications for education. British Journal of Music Education, 14(01), 19-34.
Krampe, R. T. & Ericsson, K.A. (1996). Maintaining excellence: Deliberate practice and elite performance in young and older pianists. Journal of Experiment Psychology: General, 125(4), 331-359.
Lehmann, A. C. & McPherson, G. E. (2002). Conceptions of musical ability. Research Studies in Music Education, 18(1), 19-32.
Patterson, B. (1974). Musical dynamics. Scientific American, 231(5), 78-95.
Piano World Forum (2011). https://forum.pianoworld.com/ubbthreads.php/topics/1688281/1.html
Shea, J. B., & Morgan, R. L. (1979). Contextual interference effects on the acquisition, retention, and transfer of a motor skill. Journal of Experimental Psychology: Human Learning and Memory, 5(2), 179-187.
Turner, C., Visentin, P, Oye, D., Rathwell, S., & Gongbing, S. (2021). Pursuing artful movement science in music performance: single subject motor analysis with two elite pianists. Perceptual and Motor Skills, 128(3) 1252–1274.
Varela, W., Abrami, P., & Upitis, R. (2014). Self-regulation and music learning: A systematic review. Psychology of Music, 1-20.
Wapnick, J. (1987). A comparison of tempo selections by professional editors, pianists, and harpsichordists in Bach’s Well Tempered Clavier, Book I. In C. K. Madsen and C. A. Prickett (eds.), Applications of Research in Music Behavior. University of Alabama Press, 190-203.
Zhdanov, D. (2023). A SCANDALOUS Theory about Horowitz's peculiar Piano Playing Technique. https://www.youtube.com/watch?v=C6iVT1qyj_8
Zimmerman, B. J., & Kitsantas, A. (2002). Acquiring writing revision skill: Shifting from process to outcome self-regulatory goals. Journal of Educational Psychology, 94(4), 660-668.
Appendix A: ChatGPT 3.5 Presents Effective Ways for Practicing at the Piano
Effective piano practice involves a combination of focused and strategic approaches to improve technique, musicality, and overall performance. Here are some tips for making your piano practice more effective:
1. Set Clear Goals:
Define specific, measurable goals for each practice session. Whether it's mastering a particular piece, improving a specific passage, or working on technical skills, having clear objectives helps keep your practice focused.
2. Practice Regularly:
Consistency is key. Regular, shorter practice sessions are often more effective than infrequent, lengthy ones. Aim for daily practice to build muscle memory and reinforce learning.
3. Warm-Up:
Begin each practice session with a warm-up to prepare your fingers and hands for playing. Simple scales, arpeggios, and hand-independence exercises can help increase flexibility and prevent injury.
4. Focus on Technique:
Dedicate part of your practice time to technical exercises. Work on scales, finger exercises, and other drills to improve finger strength, dexterity, and overall technique.
5. Practice Slowly:
Start practicing at a slow tempo to ensure accuracy and control. Gradually increase the speed as you become more comfortable with the passage. Playing slowly helps build muscle memory and ensures you learn the correct fingering and hand coordination.
6. Isolate Challenging Sections:
Identify the challenging sections of a piece and isolate them for focused practice. This could involve practicing a specific measure, a tricky transition, or a complex rhythm until you master it before integrating it back into the whole piece.
7. Use a Metronome:
Practice with a metronome to improve timing and rhythm. Start at a slow tempo and gradually increase it as you become more comfortable. This helps develop a steady sense of tempo and precision in your playing.
8. Practice Hands Separately:
Break down complex passages by practicing each hand separately. This allows you to focus on the specific challenges each hand presents and ensures that both hands are proficient before combining them.
9. Listen Actively:
Pay attention to the sound you produce. Listen for tone quality, dynamics, and expression. Active listening helps develop a musical ear and enhances your interpretation of the piece.
10. Record Yourself:
Use a recording device to capture your playing. Listening to recordings of your practice sessions provides valuable feedback, allowing you to identify areas that need improvement and track your progress over time.
11. Take Breaks:
Avoid long practice sessions without breaks, as fatigue can lead to diminished focus and increased likelihood of mistakes. Taking short breaks helps maintain mental freshness and physical relaxation.
12. Experiment with Different Interpretations:
Once you have mastered the technical aspects, experiment with different interpretations of the piece. Explore variations in dynamics, phrasing, and tempo to add expressiveness to your playing.
Remember that effective practice is not just about the amount of time spent but the quality of that time. Tailor your practice routine to address your specific needs and goals, and be patient and persistent in your efforts to improve.
Appendix B: ChatGPT 3.5 Presents Research Dealing With Effective Ways of Practicing at the Piano
Research on effective piano practice has explored various aspects of learning, including cognitive processes, motor skills, and pedagogical approaches. Here are some key areas of research related to effective piano practice:
1. Deliberate Practice:
The concept of deliberate practice, popularized by psychologist Anders Ericsson, emphasizes focused and purposeful practice. Studies in music education, including those specific to piano, support the idea that deliberate practice, which involves setting specific goals, receiving feedback, and repeating challenging tasks, leads to greater skill development.
Reference: Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100(3), 363-406.
2. Slow Practice and Chunking:
Research suggests that practicing at slower tempos can enhance accuracy and learning, particularly when focusing on complex passages. This aligns with the concept of "chunking," where information is organized into meaningful units. Slow practice allows for the deliberate chunking of musical information.
Reference: Hallam, S. (1997). The development of metacognition in musicians: Implications for education. British Journal of Music Education, 14(01), 19-34.
3. Motor Learning and Memory:
Studies in motor learning and memory provide insights into the physical aspects of piano playing. Research indicates that distributed practice (spacing out practice sessions over time) is more effective than massed practice (cramming). This aligns with the idea that consolidation of motor skills and memory occurs more efficiently with distributed practice.
Reference: Shea, J. B., & Morgan, R. L. (1979). Contextual interference effects on the acquisition, retention, and transfer of a motor skill. Journal of Experimental Psychology: Human Learning and Memory, 5(2), 179-187.
4. Auditory Feedback and Self-Monitoring:
Studies exploring the role of auditory feedback in piano practice suggest that musicians who actively engage in self-monitoring and adjust their playing based on auditory information demonstrate improved performance outcomes. This supports the importance of developing a keen musical ear during practice.
Reference: Lehmann, A. C., & McPherson, G. E. (2002). Conceptions of musical ability. Research Studies in Music Education, 18(1), 19-32.
5. Practice Strategies and Self-Regulation:
Research in music education emphasizes the importance of teaching students effective practice strategies and fostering self-regulation. Providing students with tools to set goals, monitor their progress, and adapt their practice strategies can contribute to more efficient learning.
Reference: Zimmerman, B. J., & Kitsantas, A. (2002). Acquiring writing revision skill: Shifting from process to outcome self-regulatory goals. Journal of Educational Psychology, 94(4), 660-668.
While these studies offer valuable insights, it's essential to recognize that individual differences play a significant role in effective practice. Tailoring practice strategies to an individual's learning style, goals, and preferences is key to optimizing the effectiveness of piano practice.
Footnotes:
[1] ChatGPT’s information appears accurate, though incomplete. It doesn’t include anything about fingering, the study of musical style, the effects of social reinforcement, the uses of imagery, and the effectiveness of practice by time of day.
[2] The timings of Alexander Scriabin’s recordings of the Poéme, though quite slow (3:30 and 3:43), are suspect, as they come from music rolls. For this reason, they are not included in Table 1.
[3] I fiddled around with trying to imbed with performance within this paper, but abandoned the idea out of frustration.
[4] I hate how I look in profile, which is why, for a time, I video-recorded myself so that my fingers were shown but my head was outside the viewing frame.
[5] This was the case for certain woodwind instruments, such as the oboe and bassoon, when performers were playing scales.
This paper was presented at the 25th Clifford K. Madsen International Symposium for Research in Music Behavior, Asheville, NC, March 20-23, 2024.
While preparing to move from Cleveland Heights to Albuquerque eighteen months ago, I dispensed with many items. Among them were tape cassettes and reel-to-reel tapes of my solo piano performances. As I hadn’t listened to them in decades and no longer owned a tape recorder, I had no justification for keeping them. I was curious to hear what I had sounded like back then, however, so before I trashed them I had them digitized. The oldest recordings dated from 1973, when I was a music performance MFA student at Sarah Lawrence College. The most recent one came from the last of three solo recitals I gave at McGill University, in 1984.
I was surprised to discover that my long held beliefs about these performances did not match what I was now hearing. The performances I remembered as bad hadn’t been so bad, and the performances I remembered as good weren’t as good as I had hoped they would be. Listening to the recordings thus gave me both relief and disappointment.
One recording interested me more than the others. It was of Scriabin’s Poéme, op. 32 no. 1, and it caught my attention for two reasons: my performance of it was good, and I hadn’t remembered playing it in recital. Certainly I recalled learning the piece, but I had thought that my only performance of it in front of others was at a 1975 audition. Yet there it was on the recording of a 1976 Syracuse University recital; I had played it as an encore.
As I listened to the Poéme a second time, I wondered if I still might be able to play that well. Krampe & Ericsson (1996) found that the average performance level of older expert pianists was only slightly lower than it was for younger expert pianists. They concluded that older pianists were able to maintain their skills because they engaged in deliberate practice throughout their lives. Given that I had kept up my practicing across the years, perhaps I too would be able to play as well as I once did.
My fingers no longer look like they used to. Six are slightly bent at their distal phalanges, and x-rays revealed that all ten are missing cartilage. I have been unable to make a tight fist in either hand since 2020 due to arthritis, and my hand span is smaller than it was when I was younger. In addition, my brain at age 77 undoubtedly looks and works differently than it did when I was 30. I haven’t noticed major negative effects on my piano playing arising from these changes, but perhaps any deterioration has been gradual or minimal enough to avoid detection. Or maybe my technique really hasn’t changed. I could have tested myself more stringently by relearning one of the more technically demanding pieces I had played long ago, say the Liszt Sonata in B Minor, but such an undertaking would have been very time-intensive. Instead, I decided to work on the three-minute Poéme. I was motivated for three reasons: first, I love the music; second, the challenge of trying to match my 1976 performance intrigued me; and third, I would use the opportunity to try some new (for me) ways of working at the piano. This last reason is my justification for presenting this case study, as I believe that what I learned from the experience has implications for both research and learning.
How I practiced
I approach a composition learned years earlier as if I had never played it. I start by practicing interlocking measures: after working on the first two measures until I can play them smoothly, I work on measures two and three, then measures three and four, and so on. I gradually increase the size of practice units to coincide with phrases, sections, and, finally, the full score. I memorize the music as quickly as I can, and I practice predominantly slowly at first, more frequently at final performance tempo later on. I also adhere to some of the ideas presented by ChatGPT in Appendices A and B, such as practicing regularly, practicing hands separately, and recording myself on occasion.[1] In summary, my way of learning and relearning piano music might be considered traditional. I suppose it also fits the definition of deliberate practice, regardless of whether the opposite of deliberate is either mindless or accidental.
My modus operandi changed in 2020. It was then that Covid-19 transformed tens of thousands of musicians into exhibitionists. Like them, I decided that what the world needed were my videoed recordings. Despite the existence of dozens of excellent professional performances on YouTube for any piece I might choose to work on, I was seduced by the possibility that posting my own performances would make a contribution.
An added appeal of my project was that I could play for others without having to experience stage fright. Though the recording process can be nerve wracking, the prospect of unlimited takes makes it bearable. I bought a Zoom Q8 “Handy Video Recorder” and a microphone stand, which along with my Steinway grand piano and living room was all I needed. I felt confident that at a minimum, my playing would improve as I worked toward a polished performance. At a maximum, I would produce recordings good enough to give others pleasure.
I changed my old practice regimen in three ways. The first followed from my purchase of the Zoom 8: I recorded myself incessantly. The second was that I collected data in order to quantify changes in my performance. By timing myself and rating my performances’ technical and musical qualities, I would strive for a series of “personal bests.” In this way I treated piano practice as much as an athletic endeavor as an artistic one.
Neither of these approaches is new. Musicians have been studying their own performances via recordings ever since home sound recording became feasible. Certainly there has been much research using both audio and video to improve almost every aspect of music learning. Both recording oneself and taking data on one’s performance can be considered examples of self-regulation in music learning (Varela et al., 2014), about which there also is much research. However, I couldn’t find any research that employed self-recording and self-data collection as monomaniacally as I did in this study.
The third and final way in which I modified my piano practice was by listening to and observing audio and video recordings of expert performances of the Poéme while I was relearning the work. This activity is controversial within the classical piano community (Piano World Forum, 2011). Some respondents feel that listening to others stunts a student’s creative development. Others insist that the activity could help one make informed decisions about interpretation, especially if the musician is exposed to many different performances rather than one. I used observation both to seek out new interpretations and to learn in what ways my own performances didn’t measure up to professional recordings.
Scriabin’s Poéme op. 32 no. 1
Faubion Bowers, author of a definitive biography of Scriabin (Bowers, 1996), describes the Poéme as follows:
In it [Scriabin] flung magical distributions of tones into that unknown nowhere he so loved, and whence he evoked such startlingly felicitous sonorities. Pleasing quietude caresses the ear here. The beginning is marked con affetto, or with affection. Sabaneeff calls the Poéme “an erotic kiss…a kind of sexual dissolving in waves of emotion.” Magic and affection are present, certainly, but where? Is it in the symmetry?
Unknown nowheres notwithstanding, one might conclude from this description that the Poéme does not require great speed or power. This is true. Instead it presents difficulties in the realms of touch, projection, dynamic control, and rhythm. The Poéme demands the ability to play cross-rhythms such as three-against-five accurately and smoothly, and dynamics must be controlled so that the melody projects while subordinate lines do not become submerged. It’s a contrapuntal work that combines passion with delicacy, and it’s not easy to pull off.
Method
Video-recording
After rememorizing the Poéme, which took about a month, I began video-recording myself. I had recorded infrequently in the past, but for the Poéme I recorded takes on about half of my practice sessions. I’d play through the piece once or twice slowly, work intensively on sections that were not secure, then record myself while playing the piece in its entirety at its final tempo from one to three times. Altogether, I recorded the Poéme about 300 times over eight months.
Record keeping
Quantifying improvement, like a runner or a swimmer keeping track of trial times, or a baseball pitcher using a speed gun to measure pitch velocity, became an important aspect of my practice. I logged timings of completed performances, numbers of mistakes made, and numerical ratings of expressivity. I didn’t statistically analyze these data because I was wary of reliability for the second and third measurements; as for the first, it was obvious from looking at the data that gradually I played the Poéme faster as I continued working on it.
Listening to other performances
I listened to other performances of the Poéme every third or fourth day. One of my most frequently used techniques was to hear in sequence a YouTube performance by a concert pianist, one of my own recent recordings, and a performance by a second concert pianist. The procedure often depressed me, but it did show me clearly what I needed to do to make my version more professional.
Results
Tempo
My first recordings were disappointing not just because I was still making mistakes, but because I was playing too slowly. I was surprised at my slowness. My tempos seemed right when I was playing, and it was only when I listened afterward that I realized how sluggish they were. This divergence between how I perceived tempo while playing versus listening persisted for months.
“Too slow” might be considered a matter of opinion. Certainly there are numerous examples on CD of performances by great pianists that are aberrantly slow. In addition, it is known that performance tempo may vary considerably across pianists (Wapnick, 1987), and even across different performances by one pianist. I had thought that my initial 2023 performances dragged, but were they slow compared to how the Poéme is played by experts? In order to find out, I compiled a list of recordings as performed by 15 select professional pianists. These were chosen from over 100 different recordings of the Poéme I found on Classical Music Online (https://classical-music-online.net) and on YouTube. I added my 1976 recital performance and two video-recordings from 2023 to the list, shown in Table 1[2]. The duration of each recording was established after locating its beginning and conclusion, as recordings sometimes started or ended with dead space or applause.
Table 1: Durations of the Scriabin Poéme on Selected Recordings
Performer Duration (minutes:seconds)
Maria Lettberg 2:22
Christopher O’Reilly 2:22
Daniil Trifonov 2:37
Yunchan Lin 2:50
Heinrich Neuhaus 2:52, 3:08
Joel Wapnick (2023) 2:52, 3:23
Joel Wapnick (1976) 2:55
Valentina Lisitsa 2:57
Vladimir Sofronitsky 3:00, 3:14
Vladimir Horowitz 3:04, 3:16
Yuja Wang 3:10, 3:48
Mikhail Pletnev 3:10
Walter Gieseking 3:12
Vladimir Ashkenazy 3:15
Josef Hofmann 3:20
Joseph Villa 3:20
Anna Fedorova 3:36
Ivo Pogorelich 4:18
With the exception of one of Yuja Wang’s two performances and Ivo Pogorelich’s outlier, my early 2023 recordings (not included in Table 1), which ranged from 3:40 to 3:50, were slower than those of the professional pianists. Indeed, I had been playing too slowly.
My 1976 and 2023 Performances of the Scriabin Poéme
I’ve posted two files of my Poéme performances on YouTube; the links to them are given below.
The 1976 performance is here:[3]
https://www.youtube.com/watch?v=83AI_jFmmFU
If you get an “Advanced Threat Protection” warning, just click the “Skip Verification” button.
My two best (in my opinion) 2023 performances are here:
https://www.youtube.com/watch?v=p2nmrKAkrMk
I’ve included them both because I couldn’t decide which I liked better. They differ considerably in tempo.
The 1976 performance, to my ears, sounds more technically assured than either 2023 performance. Dotted rhythms are played more precisely, and the differences between staccato and legato passages are more pronounced. These advantages are particularly noticeable in the Poéme’s short sixteenth-note passages, which occur eight times throughout the piece.
The sixteenth-notes are also played more softly in the 1976 recording than in the 2023 recordings.
Another impression I received is that in the 1976 recording, differences between loud and soft playing are more pronounced than in either 2023 recording. In order to verify my impression, I would have had to somehow measure loudness variability in each performance, while keeping overall loudness constant. Given the limited technology I had at my disposal, I wasn’t able to do this.
It is interesting that although the duration of the 1976 performance is three seconds longer than it is for the faster of the 2023 performances, the tempos of the first six measures of the Poéme’s two B sections (the first three lines of pages 2 and 4 of the score, respectively) are considerably slower (at 22 and 23 seconds, respectively) than they are in the 1976 performance (19 and 19 seconds, respectively). Tempo variability apparently was greater in the earlier recording.
The second 2023 performance on the video is about 31 seconds, or 17%, slower than the first 2023 performance. It sounds more relaxed than both the 1976 performance and the faster 2023 performance, and at certain times of the day I prefer it over the others. So much for listener reliability.
Discussion
Self-Recording
Though there is a considerable amount of research on audio and video self-feedback, and on self-regulation in music performance, I was unable to find studies that employed self-recording to the degree that I did in this study. I recorded myself regularly for eight months, made hundreds of recordings, and spent much time listening to them. I continue to use self-recording as part of my regular practice because I find it valuable.
Self-recording makes me self-conscious. I spend much more time evaluating my playing when I’ve recorded myself than when I haven’t. In fact I feel as if another person is in the room when the Zoom Q8 is on. Even after many months of recording myself, this feeling has not gone away. During a successful recording, I’ll get progressively more anxious the closer I am to the end of the piece. This of course makes it increasingly difficult to produce error-free recordings. On the other hand, recording oneself may be good preparation for playing in front of an audience.
Record keeping
The data I produced from record keeping was not useful, especially as it may have been unreliable. It is difficult to count mistakes while playing. My overall evaluations (on a 100-point scale) of how well I played after finishing a take may be more reliable than counting errors, but my standards may have changed from month to month. There was an important benefit of record keeping, however: it motivated me. On a day-to-day basis, I have difficulty detecting improvement. Even over longer periods this is not easy—not only because improvement at an instrument comes slowly, but because it is difficult to accurately recall how well I played, say, a month or two earlier.
Listening and watching model performances
Listening and watching recordings and videos by professional musicians made me aware of many details in the music that otherwise I might have overlooked. I adopted a few fingerings from other videos, and made note of the surprisingly frequent moments when professionals played something different from what the score indicated.
Sometimes listening to others made me aware of certain aspects of performance that I wanted to avoid. Many experts injected what I thought was too much rubato into their performances. Others omitted certain details in the music that I felt were indispensable. Musicians can become hypercritical when listening to other musicians play “their” piece, and I am no exception.
Of the three interventions, my guess is that self-recording was most influential in helping me improve. It was also the most cringe-inducing. I hated hearing when I sounded like “two elephants,” as my teacher sometimes described my hands when I played too loudly or clumsily. Even worse was when I sounded too timid. Observation helped me notice mistakes both of omission and commission, but even more important was how listening to my own performances helped me to see if I was making compelling music.
I could have recorded myself without video, but decided to include it primarily to observe how my fingers, hands, arms, and body moved. I don’t think there was any connection between these movements and how well a take sounded. Nor do I believe that there was anything problematic in my technique, as shown in the videos[4].
Performance differences
Was I able to play Scriabin’s Poéme in 2023 as well as I did in 1976? The three performances did not differ much with regard to accuracy. Moreover, the faster of the two 2023 performances was played at the same tempo and with similar rhythmic control as the 1976 performance. Nevertheless, there are a few aspects of the 1976 performance that I didn’t—or couldn’t—emulate 46 years later. Differences between staccato and legatoplaying are more evident in the 1976 recording. Moreover, the earlier recording sounds more improvisatory through its use of a wider dynamic range and greater tempo variability, and by making changes in the music suddenly rather than gradually. Consequently, I believe that the earlier recording achieves a sense of freedom and vividness that is lacking in the recent performances. On the other hand, the slower 2023 performance has a relaxed charm that is perhaps more in line with Faubion Bowers’ description of the Poéme than is the 1976 performance.
I had long known that wide dynamic range in performance contributes to expressivity. Patterson (1974), however, showed that in certain cases a pleasing effect of smoothness could be achieved by playing within a restricted dynamic range[5]. A case might be made that such smoothness is appropriate in the Poéme, given its relatively calm nature. If so, my 2023 performances may be better suited to the musical content of the Poémethan I think they are.
Limitations of the performer
I found it difficult to radically alter my playing after I had learned the Poéme well. I might listen to a take, for example, and resolve to play the following one more quickly. When I would listen afterward, confident that I had succeeded, I would discover that I had shaved off only a second or two from the earlier take. What I thought I had been doing while playing did not match what I actually did. Or I’d be satisfied with a take, thinking that it was at least as good as my 1976 recording. Then I’d listen to the 1976 recording afresh and find that my latest recording had fallen short yet again. My inability to achieve the level of playing evident in the 1976 recording was the main reason I continued practicing a three-minute piece for eight months.
Eventually I was able to match my target speed of 2:55—the duration of my 1976 performance—but it took me much so-called deliberate learning to get there. Perhaps the importance of deliberate learning lies not so much in paying attention as it does in its promotion of a willingness to experiment.
Reliability
A comparison of the artistic value in my Poéme performances must begin with the understanding that, beyond technical proficiency, preferences sometimes change with the wind. I prefer my 1976 performance of the Poéme to either of the 2023 performances most of the time, but not all of the time. I continue to experience some indecision in determining which of my recordings is best. It is possible that even for the most serious of listeners, performance preference may change—not just over months and years, but within a day or even an hour.
I don’t recall how difficult it was for me to learn the Poéme back in the mid-1970s. My guess is that relearning it last year took longer, and was more arduous. Here’s another guess: older pianists learn new pieces more slowly than do younger pianists. They may be more reluctant to learn new pieces as well.
The specific circumstance of the 1976 performance may have resulted in a freer performance than anything I could have produced, either recently or 46 years ago. I had played the Poéme at the end of a long recital, and it was preceded on its second half by two highly emotional and virtuosic pieces: Mussorgsky’s Pictures at an Exhibition and Ravel’s Alborada del gracioso. When it was time to play the Poéme, I may have been unusually inspired and relaxed. Such a potentially idiosyncratic situation shows how my study is subject to the weaknesses of many case studies, in that circumstances may have been unique enough to make the validity of generalizations from it suspect.
Problems of cause and effect
One might consider self-recording, record keeping, and listening and watching model performances to be the three independent variables in this case study. A problem with this formulation is that there is no control group. Indeed, the only real “treatment” here is the passage of time. Musicians change in performance proficiency for many reasons, including physical and cognitive aging, increased knowledge of music, aesthetic development, the influences of other musicians, and changes in the amount of practice. If performances were not to change much over many years, it is unclear whether we should marvel at how resistant they are to the effects of time, or wonder about the musician’s lack of artistic growth.
It would be interesting to gain a greater sense of how musical performance changes with age. There appear to be wide differences between individuals. Most of us have witnessed in concert both musicians who should have quit years ago and musicians who appear to be ageless, for example. Also, knowledge of what aspects of musical technique and expressiveness are most susceptible to the effects of time has not been extensively researched. Knowledge in this respect might be gained by having older accomplished musicians relearn music that they recorded long ago but haven’t practiced recently.
In certain sports (swimming, gymnastics, tennis), peak performance comes early. For musicians, this may or may not be the case. Pianists such as Artur Rubinstein, Vladimir Horowitz, Claudio Arrau, and Menachem Pressler pleased audiences well into their 80s (90s in the case of Pressler). On the other hand, at a purely technical level these pianists were far superior when they were in their 40s. If in fact their later performances can be shown to be more musically satisfying than their earlier ones, it would be interesting to know how the older masters accomplished such miracles. Is superior late-life performance a consequence of having experienced more of life, and of music? Or is this supposition false, and the goodness of a musical performance primarily depends on the performer’s technique?
Implications
The literature on playing the piano is enormous. Hundreds of books and studies have been written on technique, piano teaching, historical performance practice, and practical advice for pianists of all kinds and at all levels. More are filled with preparatory exercises written by the likes of Cramer, Hanon, Czerny, John Thomson, and countless others—to say nothing of technical exercises and etudes—real music!—composed by Bach, Clementi, Chopin, Liszt, and Brahms.
With fecundity comes disagreement, and with disagreement comes the need for answers. Perhaps the most obvious controversy in piano performance relates to technique. Should a pianist play with curved or flat fingers (Zhdanov, 2023)? High or low wrist? How much involvement of the forearms, upper arms, and entire upper body is ideal?
One might question the implied either-or dichotomy inherent in these questions. In sports, there is considerable variability in certain well practiced motions such as tennis serve, baseball pitching and batting, golf swings, and basketball foul shooting. Similarly, it is possible that there is no one right way of playing the piano. Or maybe there is? If we look to experimental research on the topic, we find little to guide us other than motion studies. A partial bibliography of such studies (Turner et. al, 2021), reveals that this research consists primarily of case studies—alike in form though not in the specifics of what I’ve done in this case study.
When it comes to researching how to play the piano, there may be little alternative to the case study. We cannot randomly assign students to treatments in order to answer even basic questions such as whether practicing scales versus appropriate control activities is beneficial, what warmup exercises and technical approaches are most resistant to injury, and whether practicing beyond three, six, or nine hours a day has long-term negative consequences. Indeed, almost all of ChatGPT’s recommendations in Appendix A have come down to us as given truths learned through extensive experience. There is value in them—probably. On the other hand, I haven’t used a metronome in fifty years. What have I been missing?
In conclusion, one case study about piano playing demonstrates nothing. But maybe ten, or a hundred, can.
References
Bowers, Faubion (1996). Scriabin: a Biography. Second edition, Mineola, New York: Dover Publications.
Ericsson, K.A., Krampe, R.T., & Tesch- Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100(3), 363-406.
Hallam, S. (1997). The development of metacognition in musicians: Implications for education. British Journal of Music Education, 14(01), 19-34.
Krampe, R. T. & Ericsson, K.A. (1996). Maintaining excellence: Deliberate practice and elite performance in young and older pianists. Journal of Experiment Psychology: General, 125(4), 331-359.
Lehmann, A. C. & McPherson, G. E. (2002). Conceptions of musical ability. Research Studies in Music Education, 18(1), 19-32.
Patterson, B. (1974). Musical dynamics. Scientific American, 231(5), 78-95.
Piano World Forum (2011). https://forum.pianoworld.com/ubbthreads.php/topics/1688281/1.html
Shea, J. B., & Morgan, R. L. (1979). Contextual interference effects on the acquisition, retention, and transfer of a motor skill. Journal of Experimental Psychology: Human Learning and Memory, 5(2), 179-187.
Turner, C., Visentin, P, Oye, D., Rathwell, S., & Gongbing, S. (2021). Pursuing artful movement science in music performance: single subject motor analysis with two elite pianists. Perceptual and Motor Skills, 128(3) 1252–1274.
Varela, W., Abrami, P., & Upitis, R. (2014). Self-regulation and music learning: A systematic review. Psychology of Music, 1-20.
Wapnick, J. (1987). A comparison of tempo selections by professional editors, pianists, and harpsichordists in Bach’s Well Tempered Clavier, Book I. In C. K. Madsen and C. A. Prickett (eds.), Applications of Research in Music Behavior. University of Alabama Press, 190-203.
Zhdanov, D. (2023). A SCANDALOUS Theory about Horowitz's peculiar Piano Playing Technique. https://www.youtube.com/watch?v=C6iVT1qyj_8
Zimmerman, B. J., & Kitsantas, A. (2002). Acquiring writing revision skill: Shifting from process to outcome self-regulatory goals. Journal of Educational Psychology, 94(4), 660-668.
Appendix A: ChatGPT 3.5 Presents Effective Ways for Practicing at the Piano
Effective piano practice involves a combination of focused and strategic approaches to improve technique, musicality, and overall performance. Here are some tips for making your piano practice more effective:
1. Set Clear Goals:
Define specific, measurable goals for each practice session. Whether it's mastering a particular piece, improving a specific passage, or working on technical skills, having clear objectives helps keep your practice focused.
2. Practice Regularly:
Consistency is key. Regular, shorter practice sessions are often more effective than infrequent, lengthy ones. Aim for daily practice to build muscle memory and reinforce learning.
3. Warm-Up:
Begin each practice session with a warm-up to prepare your fingers and hands for playing. Simple scales, arpeggios, and hand-independence exercises can help increase flexibility and prevent injury.
4. Focus on Technique:
Dedicate part of your practice time to technical exercises. Work on scales, finger exercises, and other drills to improve finger strength, dexterity, and overall technique.
5. Practice Slowly:
Start practicing at a slow tempo to ensure accuracy and control. Gradually increase the speed as you become more comfortable with the passage. Playing slowly helps build muscle memory and ensures you learn the correct fingering and hand coordination.
6. Isolate Challenging Sections:
Identify the challenging sections of a piece and isolate them for focused practice. This could involve practicing a specific measure, a tricky transition, or a complex rhythm until you master it before integrating it back into the whole piece.
7. Use a Metronome:
Practice with a metronome to improve timing and rhythm. Start at a slow tempo and gradually increase it as you become more comfortable. This helps develop a steady sense of tempo and precision in your playing.
8. Practice Hands Separately:
Break down complex passages by practicing each hand separately. This allows you to focus on the specific challenges each hand presents and ensures that both hands are proficient before combining them.
9. Listen Actively:
Pay attention to the sound you produce. Listen for tone quality, dynamics, and expression. Active listening helps develop a musical ear and enhances your interpretation of the piece.
10. Record Yourself:
Use a recording device to capture your playing. Listening to recordings of your practice sessions provides valuable feedback, allowing you to identify areas that need improvement and track your progress over time.
11. Take Breaks:
Avoid long practice sessions without breaks, as fatigue can lead to diminished focus and increased likelihood of mistakes. Taking short breaks helps maintain mental freshness and physical relaxation.
12. Experiment with Different Interpretations:
Once you have mastered the technical aspects, experiment with different interpretations of the piece. Explore variations in dynamics, phrasing, and tempo to add expressiveness to your playing.
Remember that effective practice is not just about the amount of time spent but the quality of that time. Tailor your practice routine to address your specific needs and goals, and be patient and persistent in your efforts to improve.
Appendix B: ChatGPT 3.5 Presents Research Dealing With Effective Ways of Practicing at the Piano
Research on effective piano practice has explored various aspects of learning, including cognitive processes, motor skills, and pedagogical approaches. Here are some key areas of research related to effective piano practice:
1. Deliberate Practice:
The concept of deliberate practice, popularized by psychologist Anders Ericsson, emphasizes focused and purposeful practice. Studies in music education, including those specific to piano, support the idea that deliberate practice, which involves setting specific goals, receiving feedback, and repeating challenging tasks, leads to greater skill development.
Reference: Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100(3), 363-406.
2. Slow Practice and Chunking:
Research suggests that practicing at slower tempos can enhance accuracy and learning, particularly when focusing on complex passages. This aligns with the concept of "chunking," where information is organized into meaningful units. Slow practice allows for the deliberate chunking of musical information.
Reference: Hallam, S. (1997). The development of metacognition in musicians: Implications for education. British Journal of Music Education, 14(01), 19-34.
3. Motor Learning and Memory:
Studies in motor learning and memory provide insights into the physical aspects of piano playing. Research indicates that distributed practice (spacing out practice sessions over time) is more effective than massed practice (cramming). This aligns with the idea that consolidation of motor skills and memory occurs more efficiently with distributed practice.
Reference: Shea, J. B., & Morgan, R. L. (1979). Contextual interference effects on the acquisition, retention, and transfer of a motor skill. Journal of Experimental Psychology: Human Learning and Memory, 5(2), 179-187.
4. Auditory Feedback and Self-Monitoring:
Studies exploring the role of auditory feedback in piano practice suggest that musicians who actively engage in self-monitoring and adjust their playing based on auditory information demonstrate improved performance outcomes. This supports the importance of developing a keen musical ear during practice.
Reference: Lehmann, A. C., & McPherson, G. E. (2002). Conceptions of musical ability. Research Studies in Music Education, 18(1), 19-32.
5. Practice Strategies and Self-Regulation:
Research in music education emphasizes the importance of teaching students effective practice strategies and fostering self-regulation. Providing students with tools to set goals, monitor their progress, and adapt their practice strategies can contribute to more efficient learning.
Reference: Zimmerman, B. J., & Kitsantas, A. (2002). Acquiring writing revision skill: Shifting from process to outcome self-regulatory goals. Journal of Educational Psychology, 94(4), 660-668.
While these studies offer valuable insights, it's essential to recognize that individual differences play a significant role in effective practice. Tailoring practice strategies to an individual's learning style, goals, and preferences is key to optimizing the effectiveness of piano practice.
Footnotes:
[1] ChatGPT’s information appears accurate, though incomplete. It doesn’t include anything about fingering, the study of musical style, the effects of social reinforcement, the uses of imagery, and the effectiveness of practice by time of day.
[2] The timings of Alexander Scriabin’s recordings of the Poéme, though quite slow (3:30 and 3:43), are suspect, as they come from music rolls. For this reason, they are not included in Table 1.
[3] I fiddled around with trying to imbed with performance within this paper, but abandoned the idea out of frustration.
[4] I hate how I look in profile, which is why, for a time, I video-recorded myself so that my fingers were shown but my head was outside the viewing frame.
[5] This was the case for certain woodwind instruments, such as the oboe and bassoon, when performers were playing scales.