Embrace the Struggle

When I did my research project about learning and practicing music, I learned that practicing encourages the growth of layers of myelin around neural circuits. The myelination response is targeted to the neurons involved in the activity that you practice, activated by repetition and by struggle. After the project I continued to read about this topic, and really enjoyed the book, “Make It Stick” by Peter C. Brown. He explains that learning is deeper and more durable when it is effortful. When learning is harder and slower it doesn’t feel as good, but when you work hard to recall a memory, you actually strengthen it. High school and university students like highlighting and rereading textbooks, which is easy and feels good and gives them the illusion of knowing something. But testing yourself, practicing retrieval, while it feels much slower and more difficult, is demonstrably more effective at strengthening the memory. The act of retrieving a memory changes the memory and makes it easier to recall it again.

This is also applicable to the use of interleaved practice versus massed or blocked practice, which I first learned about on The Bulletproof Musician blog site and discussed in my project writeup. There is ample research in various kinds of learning that shows that the increased effort needed to remember after forgetting actually strengthens the memory. You don’t want so much forgetting that you have to relearn the material, just enough to increase the effort and struggle. But the forgetting is actually essential for the learning.

I have found that tunes that I learn easily and quickly during class are quickly forgotten. Tunes that require more effort are easier to retain. Sometimes the technically difficult part that I had to practice more is the part that I remember best. Also, if I learn something from sheet music, my brain knows it is there on the sheet and doesn’t work to keep the memory. Music I have learned by ear is much more likely to stick. I work to repeat new tunes every day or so until I convert them to long-term memory. First I play the recording and play along. After the tune memory gets stronger I play a few notes of the recording and then play the tune on my own. Finally I am able to think of the whole tune on my own from the name. I find that I can do this easily with some tunes, but that others take more work to recall. I practice recalling tunes from the list of names, because now I know that this activity strengthens the memory and facilitates the retrieval of the memory.

Key points are that:

  • Difficulty is a key part of learning and can be beneficial
  • Errors are natural and to be expected
  • Failure is an essential experience on the path to mastery
  • Practice helps

Make it Stick cover

Capturing tunes

With the course almost over, we now have 166 tunes we have been given at the ESI; 88 of these are primary tunes that we are expected to learn. I excluded songs from this list, although some of these instrumental tunes also have songs.

Two years ago I started making a list of tunes I had learned. I got the idea from Sheila Morris. We sat down to jam together after a Seattle Springdans workshop and Sheila handed me list of tunes she knew. It made it much easier for us to find tunes in common to play. After that I started my own list. I soon migrated to an Excel spreadsheet; as a scientist in my working life I was very comfortable with Excel. I included columns to mark which tunes I could play alone without too much embarrassment, which tunes I could play with others, which tunes I was only just learning. I noted which ones had sheet music (in my possession) and which had recordings, and the filenames to find the recordings or sheet music. Before I came to the ESI this list had grown to 175 tunes. I use various features of Excel, such as filters to list just tunes I know well, or make a setlist for playing for dance. I can also filter by type of tune, such as Schottis or Bodapolska.

When the tunes started to pile up here at the ESI, I began a worksheet to capture those tunes as well. I am really happy I did that, because it helped me keep track and to make lists of tunes I needed to practice. This week it helped me put together a list of the tunes we are choosing for our upcoming final concert.

Here is an example of a section of this worksheet. It includes our four main teachers and visiting teachers David Eriksson and Markus Svensson:

Dance Tune Name Efter learned from date 1st recorded or learned
32 Polska Hem från Gesunda Sonia Sahlström 19/09/14
33 Polska Polska efter Mårten Blank, den vanliga Mårten Blank Sonia Sahlström 19/09/14
34 Polska Runnom e. Johan Hollsäter, Flemdalen, Norge Johan Hollsäter Mia Marin 22/09/14
35 Menuett Menuett e. Gustav Blidström, Skara Gustav Blidström Mia Marin 22/09/14
36 Polska Polska efter Olof Andersson, Eda, Värmland Olof Andersson Mia Marin 22/09/14
37 Slängpolska Slängpolska e. Ola Lantz, Skåne Ola Lantz Mia Marin 22/09/14
38 Slängpolska Lasse Leila Diu Gunnel Maritzson, Gotland Mia Marin 22/09/14
39 Polonaise Polonaise e. Gustav Blidström, Skara Mia Marin 22/09/14
40 Vals Rapp Kalle alt springvals Rapp Kalle Sonia Sahlström 26/09/14
41 Vals Bakvände Reprisevalsen Gås Anders Sonia Sahlström 26/09/14
42 Bondpolska från Viksta Gästrikepolska Per Persson Menlös Olov Johansson 08/10/14
43 Polska Polska från Munkedal David Eriksson 04/11/14
44 Polska Polska efter Alfred Nilsson David Eriksson 04/11/14
45 Vals Salbohedsvalsen Västmanland David Eriksson 04/11/14
46 Vals Vals fran Vestmanland David Eriksson 04/11/14
47 Finskoog’s pols Fanteladda Olov Johansson 05/11/14
48 Polska Polska från Eda Olov Johansson 05/11/14
49 Bondpolska från Viksta Första Gangen som det var lyst Viksta-Lasse Ditte Andersson 06/11/14
50 Bingsjö Kaker brö efter Viksta-Lasse Hjort Anders Ditte Andersson 06/11/14
51 Polska Sågfallet efter Oskar Larsson Oskar Larsson Ditte Andersson 06/11/14
52 Polska Jul igen från Risinge, Östergötland Ditte Andersson 06/11/14
53 Polska Gubbandansen Markus Svensson 07/11/14
54 Polska Här_dansar_jag and song Markus Svensson 07/11/14
55 Polska Polska_fr_Dalskog Markus Svensson 07/11/14

 

Project: Scientific Studies About Practicing Music, part 2

My training as a scientist prompted me to wonder how optimised musical practice improves the motors skills needed for expert music performance. I read about the mechanism of learning in the brain. There has been significant progress in recent years in understanding learning and memory, finding that myelin is highly important for learning motor skills such as those needed by musicians to produce music.

Motor skills (such as those used by musicians) are created by chains of nerve fibers carrying an electrical impulse to the muscles needed to make the music. For many years the focus of neurobiologists was on the nerve cells, the axons and the synapses that join the cells (Figure 3). Myelin is an electrically insulating material composed of 40% water, and dry mass including variable percentages of lipids (70-85%) and protein (15-30%), that wraps around the nerve fibers (Figures 3 and 4, below). For many years myelin was thought to be less important than the nerve cells themselves, but the myelin wrapping makes the signal stronger and faster and prevents the electrical impulses from leaking out. Recent studies into the molecular biology of the neuromuscular system have shown that myelin is much more important than originally thought.

Complete_neuron_cell_diagram

Figure 3. Structure of a typical neuron.

Learning new motor skills alters the structure of the brain’s white matter, made up of oligodendrocytes (OLs), and the oligodendrocyte cells produce myelin that lays down layers around the specific circuits that are engaged during motor learning. Each layer adds skill and speed. Practicing an instrument adds new layers of myelin around the correct neural circuits (Bengtsson, Nagy, Skare, Forsman, Forssberg, & Ullen, 2005). The phrase “muscle memory” is talking about these circuits. Neural traffic at 2 mph can accelerate to 200 mph with more myelin wrapping around the nerve fiber. This understanding of the science of myelin and its role in learning is relatively new, dating from around 2006.

AxonMyelin

Figure 4. This is an electron micrograph of a cross section of a nerve fiber through the axon, showing the layers of myelin wrapped around (Hubel).

 

 

 

 

 

 

 

 

It has been known for many years that most oligodendrocytes develop in the first weeks of life in mammals. It is a recent finding that oligodendrocyte formation and myelination continues throughout adulthood in the healthy adult brain. Recent experiments created genetically engineered mice in which their ability to form new oligodendrocytes could be removed during adulthood by administration of a drug, so the timing could be controlled. It was found that when mice did not have the ability to form new oligodendrocytes, they failed to learn an effective strategy in a complex task (running wheel with uneven bars). If the mice had learned the skill before losing their ability to make new oligodendrocytes, they performed normally, showing that new oligodendrocytes are not needed for information retrieval, but for learning new motor behavior (McKenzie, et al., 2014).

Additional observations about myelin: Einstein’s brain had an average number of neurons, but more glial cells, which produce and support myelin. It is only recently that that observation has made sense. Myelin deficiencies are linked to disorders such as autism, attention deficit disorder (ADD), and post-traumatic stress disorder (PTSD). Multiple sclerosis is a demyelinating disease. Age matters, myelin grows more in children. There is a net gain of myelin until the age of 50 when the balance tips towards loss. Throughout life 5 percent of oligodendrocytes remain immature and ready to answer the call, but it takes more time and sweat to build the circuitry in older adults. George Bartzokis, professor of neurology at UCLA, takes DHA fatty acids to stimulate his own myelin growth. Also, myelin wraps, but doesn’t unwrap. This is why habits are hard to break; they can only be replaced by new habits.

Now that we know that we need to encourage the growth of layers of myelin around neural circuits needed to produce music, how do we encourage our brains to develop in this way? The myelination response is a flexible and responsive system, and gives us the potential to earn skill where we need it. The key finding is that myelin responds to what you do. It is important to fire the correct circuits by repetition, but also by struggle. The strategy is to analyze and work over something you want to learn — to stumble and stop, figure it out, and then repeat correctly. The trick is to choose a goal just beyond your present abilities, to target the struggle. This is called “deliberate practice” or “deep practice” (Coyle, 2009).

Concluding remarks

I have synthesized this into a strategy with the following elements:

  1. My practices are divided into segments as suggested by Mia Marin: body warm-up, technical exercises, new tune material, and review tune material.
  2. I have adopted Markus Svensson’s suggested technical warm-up exercises that use 5-note scales on each string, first slowly (I use a ‘tonstarter’ approach to make the best possible sound), then doubling the speed with slurred notes, then separate bow strokes, repeating until it is very rapid.
  3. I use a “deep practice” strategy on new tunes in which I make sure to observe errors and stop and correct and work over them. I create small exercises to target specific technical skills as needed for the tunes I am working on. I avoid repeating something incorrectly, and repeat correctly.
  4. I also push on the speed and play at the edge of what I can do, to find the weak places and focus on those.
  5. I use an interleaved strategy in which I work on a series of tunes, with more infrequent blocks of extensive practice on a single tune. I record myself and listen, both for the feedback it provides, and to practice a performance situation where my focus is not on noting my mistakes (the recorder will hear those!).
  6. I use a practice notebook to record information about practice goals and strategies, as well as observations and notes. I take note of improvements, to build confidence.

References

Bengtsson, S. L., Nagy, Z., Skare, S., Forsman, L., Forssberg, H., & Ullen, F. (2005). Extensive piano practicing has regionally specific effects on white matter development. Nature Neuroscience , 8 (9), 1148-1150.

Coyle, D. (2009). The Talent Code. New York, New York, USA: Random House, Inc.

Duke, R. A., Simmons, A. L., & Cash, C. D. (2009). It’s not how much; it’s how — Characteristics of practice behavior and retention of performance skills. Journal of Research in Music Education , 56 (4), 310-321.

Ericsson, K. A., Krampe, R. T., & Tesch-Romer, C. (1992). The role of deliberate practice in the acquisition of expert performance. Psychological Review , 100 (3), 363-406.

Hambrick, D. Z., Altmann, E. M., Oswald, F. L., Meinze, E. J., Gobet, F., & Campitelli, G. (2014). Accounting for expert performance; The devil is in the details. Intelligence , 45, 112-114.

Hambrick, D. Z., Oswald, F. L., Altmann, E. M., Meinz, E. J., Gobet, F., & Campitelli, G. (2014). Deliberate practice: Is that all it takes to become an expert? Intelligence , 45, 34-45.

Hubel, D. (n.d.). Retrieved March 24, 2015, from Web Page about David Hubel’s book, Eye, Brain, and Vision: http://hubel.med.harvard.edu/book/b5.htm

Kagayama, N. (n.d.). Retrieved January 15, 2015, from Bulletproof musician: http://www.bulletproofmusician.com/why-the-progress-in-the-practice-room-seems-to-disappear-overnight/

Kagayama, N. (2015). Retrieved January 15, 2015, from Bulletproof musician: http://www.bulletproofmusician.com/why-the-way-we-usually-practice-makes-us-think-were-better-prepared-than-we-really-are/

Long, P., & Corfas, G. (2014). To learn is to myelinate. Science , 346 (6207), 298-299.

McKenzie, I. A., Oyahon, D., Li, H., Paes de Faria, J., Emery, B., Tohyama, K., et al. (2014). Motor skill learning requires active central myelination. Science , 346 (6207), 318-322.

Project: Scientific Studies About Practicing Music

Each music student at the ESI does a project during the year-long course. It is expected to take about 30 hours, for example 3 hours a week for about 10 weeks between January and March. The choice of topics is up to the student; many do a profile of a folk musician, or tunes from a particular region. We produce a written paper of 5-8 pages, give a 20 minute talk, and play 4-5 tunes solo for the class and teacher.

This is the first of two blog posts concerning what I learned while researching this project.

Introduction and Aim

I am devoting these 9 months in the 2014-2015 school year to the intensive study of Swedish folk music at the Eric Sahlström Institute (ESI), and this activity includes a lot of individual practice. I was interested in understanding how to maximize the effectiveness of this musical practice time, and began investigating scientific studies that addressed this issue. Although most studies concerned classical music, I applied these strategies to my folk music practice. I will here describe the studies I found most compelling, and outline the strategy I have synthesized as a result.

Methods, Results, and Discussions

First Study:

Anders Ericsson is a Swedish psychologist who spent his career trying to understand how people become expert performers in sports or music. In 1992 Ericsson, Krampe, and Tesch-Römer published a study stating that anyone can master a complex skill with at least 10,000 hours of deliberate practice. Ericsson defined deliberate practice as “engagement in highly structured activities that are created specifically to improve performance in a domain through immediate feedback, that require a high level of concentration, and that are not inherently enjoyable.” In the original study, the researchers analyzed the self-reported practice duration of three groups of age and gender-matched musicians, 1) “the best violin students” nominated by professors at the Music Academy of West Berlin (Hochs-chule der Kuenste) as students who had the potential for careers as international soloists, 2) “good violinists” at the same institution, and 3) music education students in a department with lower admissions standards, the “music teachers”. These groups of musicians were interviewed and asked about their time spent on music practice and other activities. The data were analyzed and the study concluded that elite musicians’ total practice time was much more than the other groups of musicians. The “best” violinists had spent over 10,000 hours of deliberate practice by age 20, about 2,500 h more than “good” violinists, and 5,000 h more than the average for the “teacher” group. Ericsson claimed that these high levels of deliberate practice were not only necessary, but also sufficient for exceptional performance (Ericsson, Krampe, & Tesch-Romer, 1992). This was further popularized by Malcolm Gladwell‘s book Outliers and often (over)simplified to state that anyone can master a skill with 10,000 hours of practice. This meritocratic idea has become very popular, and fueled the ongoing “nature vs. nurture” debate. However, this study has been criticized because even though there was extensive statistical analysis, the study was based on self-reported data from interviews, which can be unreliable.

Second Study:

A more recent study by a group of psychologists from five universities ((Hambrick, Oswald, Altmann, Meinz, Gobet, & Campitelli, 2014) reanalyzed expertise research on both chess and music. They examined data from 6 chess studies and 8 music studies that asked similar questions about recalled amounts of practice, but these studies also had separate metrics for skill levels so they could look for correlation between practice time and skill level. These analyses found that individual differences in accumulated amount of deliberate practice accounted for only about one-third of the reliable variance in performance levels in chess players and musicians, “leaving the majority of the reliable variance unexplained and potentially explainable by other factors.” There are many instances of people who did reach an elite level of performance without copious practice, and other people who failed to do so despite copious practice. There is a substantial amount of evidence that while deliberate practice may be necessary, it cannot be claimed to be sufficient. Other factors could be such elements as personality, the age you started, intelligence, or something else entirely, and the authors discuss these in the publication (Hambrick, Altmann, Oswald, Meinze, Gobet, & Campitelli, 2014).

Third Study:
Another compelling and informative study was done by Robert Duke at the University of Texas and published in 2009 in the Journal of Research in Music Education (Duke, Simmons, & Cash, 2009). Seventeen piano students agreed to participate in the study, practicing and then performing a 3-measure passage from Shostakovich’s Piano Concerto No. 1. This passage had some tricky elements, making it too difficult to sight-read, but was not too challenging to master in a single practice session.

Study Method: Students were allowed a two-minute warm-up, then given the sheet music, a metronome and a pencil. Students were allowed a single practice session of any duration, and chose times as short as 8 ½ minutes or long as 57 minutes. Videos and MIDI-keyboard data were collected for all practice sessions and performances for later analysis. They were asked not to practice it overnight and returned the next day for testing. The next day pianists were allowed a 2-minute warm-up but asked not to play the excerpt. They were then told to play the passage 15 times through for the study, pausing briefly between repetitions. A metronome was used initially to establish the target tempo of 120 beats per minute (bpm), but not for all 15 trials. Trials were recorded on videotape for analysis, and MIDI keyboard data were collected. Each trial was scored for correct or near-correct or incorrect performance by multiple judges, and participants were ranked from 1-17 by those judges.

Results:   Three of the 17 participants performed significantly better than the others, with more even tone, fluid execution, musical expression, and rhythmic precision. Practice data from these three high level performers were further analyzed to determine what practice strategies contributed to their better performances. Interestingly, practice time and number of repetitions of the passage did not correlate with better performance. What did correlate were specific elements of the quality and strategies of practice. The best practice strategies included a thoughtful approach, in which they located errors and worked to eliminate them. Sections were repeated incorrectly less often, and played correctly more times. The top performers slowed down at key moments to avoid repeating passages incorrectly. Minimizing the number of incorrect repetitions was key, as the lower-ranking pianists played the passage incorrectly more times. The percentage of correct repetitions out of the total number of repetitions was significantly correlated with better performance. This fits the definition of “deep practice” or “deliberate practice”.

This is something our teachers have told us at the ESI, to be sure not to repeat something incorrectly, but to fix it and then repeat it correctly multiple times.

Another informative suggestion:

Another element of practice that I investigated was the observation that often musicians feel well prepared during practice, but don’t execute as well during performance. This was ascribed to the usual method of practicing with multiple repetitions of a piece during practice until it begins to feel and sound better, a so-called blocked practice strategy, before moving on to another piece. It begins to feel satisfying to see that progress during practice, since the piece becomes better upon repetition in these sessions, but sometimes with this strategy the performance is not as good as the practice sessions. This may be because the blocked practice strategy is different from the situation in performance, which is not done with multiple repetitions. An alternative strategy is random or interleaved practice, such as playing pieces through once and going on to another. This strategy may feel initially less satisfying in the practice room, but with enough repetition leads to better performance (Kagayama, Bulletproof musician, 2015).

Another informative suggestion:

I consulted with Andrea Larson, who took the ESI music course two years ago, and now offers workshops in the USA that feature a focus on practice strategies. I had a Skype session with Andrea to ask about her approach. I will limit the detail I describe in this section and encourage people to take one of Andrea’s workshops. Andrea advocates recording yourself, and making notes in a practice notebook as many musicians do, including our teachers and a website I consulted, www.bulletproofmusician.com.

To be continued in the next blog post…

References are also listed in the next blog post…

Learning tunes by ear — technology

Our tradition is to learn all tunes by ear, no sheet music. We are learning many tunes very quickly (we have 140 so far in the course). Learning by ear is cognitively very different from reading sheet music. When I first began nyckelharpa and folk music (9 years ago) I found it challenging. But with practice it got easier. I also had some advice from Toby Weinberg, when I was complaining that I quickly forgot tunes I had just learned. He said that the forgetting is part of the learning process. You sleep and forget and then learn the tune again, and that this repetition is important. It helped me feel as though I wasn’t uniquely unsuited to do this, and was very reassuring. I think you are convincing your long term memory to keep this information stored. I learned from Olov Johansson that he heard from Bjorn Ståbi that, “you need to forget a tune three times before you really know it!”

The key to this approach is audio recordings of the tunes. I have developed some technology strategies that work for me, after several false starts.

(Disclaimers: This post is a bit mundane, but could be helpful to some readers. These are just my opinions. I have no commercial interest in any of the products mentioned here, nor any connection.)

Voice memo on the iPhone: This can sound a bit muddy, but is always handy. You have the opportunity to name the tune right after recording. However, our teachers often record a series of tunes in a row, moving on the next rather quickly, without enough time to type a name on an iPhone screen before the next recording needs to start. Or the teacher will write up the tune names later, so you don’t know how to spell it until later. I also had trouble getting voice memos to import into the Amazing Slow Downer. One student used a workaround, emailing them to herself and then putting into iTunes, and then in the Amazing Slow Downer, but I had trouble with this.

I also tried an app called Voice Recorder HD for Audio Recording eFUSION Co, Ltd from iTunes.  It can support different recording quality levels and makes .wav files. I did like the feature that you can easily edit the tune. This was useful because we sometimes start recording and then have to wait for someone, or the teacher decides to tune or talk some more before beginning the tune. These tunes were also easy to import into the Slow Downer. However, overall I hated it. It gives each tune a long name including the date and series of numbers and dashes. You can rename them, but what a pain.

What I continue to use and like best is an Olympus digital recording device; these are intended for voice recording. Calling them voice recorders may be like a disclaimer, so you don’t expect high fidelity musical quality. There are many similar devices, some are bigger and nicer. I have a VN-702PC which is small and portable. I got a microphone on a cord to plug into the Mic outlet, and I think this adds to the quality of the recording. I later attach it to my computer via USB port and drag and drop the files into a folder. They come in with filenames using sequential numbers such as 702_0433.mp3, and I add the tune names I wrote down in class just after the number, and double click to bring into iTunes with the names. These are pretty good quality, and import well into the Amazing Slow Downer.

Our class has a shared Dropbox and we upload and download to share tunes with classmates who have no recording capability, or are out of class that day.

The Amazing Slow Downer is a really valuable tool. It is software available online, or an app for the iPhone or iPad. You can import a recorded tune, and slow it down while keeping the pitch constant. It also has the ability to replay a small segment of the tune that you set, which is helpful for going over and over a hard part, or just the A part, or whatever. You can slow it down to an extreme degree and really hear the ornaments in a tune.

We have made tune transcriptions as part of our theory and arrangement education, and I have done it before as well. Sometimes I use pencil on music paper, but also I have been using Finale Print Music software. It is clunky but I am becoming more capable with it. I like being able to adjust the fit of the bars on the staffs, add parts, transpose, and hear the parts play back on the computer when I am making accompaniments and second voices. I have been making some additional transcriptions on tunes that I want to make sure to keep. If nothing else, transcribing to sheet music places the elements of the tune together with its name in a findable format, which I find useful. I have also used it for tricky or nonstandard bowing patterns that were some trouble to learn, so I can capture them, or to help my brain process them better. I use ForScore software to have pdfs of sheet music in a library or setlist on my iPad, so they can be captured there or just in a folder of files. I find this preferable to printing them out; I don’t want to import heavy paper into the United States when I go home.

IMG_0612

Our class is recording a teacher playing a tune.

DSC_0003

This is my Olympus recorder, with microphone and USB cable.