Some musicians like to listen to recordings of a piece before learning it. Others like to learn pieces from the score first. Is one approach better than the other?
Let’s say you have to learn a few short pieces in the next 24 hours, and have 4 hours to practice. How would you approach this task?
Would you just dive right in and spend all 4 hours reading through the parts and working out the details? Or would you hop on YouTube and spend 30-60 minutes listening to some recordings first before doing any work on your instrument?
As a young Suzuki kid, it was always the latter. I did a ton of listening – upon awakening, during meals, in the car, and even as I was falling asleep every night. I remember well the little Sony stereo set on my dresser that my mom would turn on and leave running as she left the room after tucking me in.
I can’t say for sure that it’s because of all the listening I did, but I moved through repertoire pretty quickly as a little kid (except for the seeming eternity I spent on Corelli’s LaFolia).
Even in my grad school days, listening to recordings was always a key part of my learning process. Whether it was a concerto or piano trio, I always sought out recordings before taking a look at the score or trying to play through it on my violin.
My wife on the other hand (a pianist), was always a look-at-the-score-first kind of person, and didn’t like listening to recordings until after she had already gotten the piece into her fingers. Nevertheless, a few months ago, when faced with a lot of rep to learn for a performance she had to do on short notice, I suggested that she put recordings of every piece on her phone and spend her 16-hr plane ride immersed in an endlessly repeating playlist.
When we spoke about it later, she said that it seemed to help her jump right into rehearsals the next day, and get everything into her fingers more quickly, jet-lag and all (though it’s possible she said this to get me to shut up about it already).
So this leads to an interesting question about the impact of listening on our learning. Is this just an old Suzuki habit that I never let go of? Or does listening to music in the early going really accelerate the learning process?
12 minutes. 13 notes.
To see what effect an auditory “model” might have on musicians’ learning rates, a team of researchers recruited 32 instrumental music majors with some secondary piano training (i.e. no more than a few years of private instruction/group piano class) to learn a short passage either (a) after listening to a recording or (b) without listening to a recording.
All students were allowed one 12-minute training session sometime between 8-10pm, in which they practiced the following 13-note melody with their left hand (all participants were righties).
From Cash, C.D., Allen, S.E., Simmons, A.L., & Duke, R.A. (2014). Effects Of Model Performances On Music Skill Acquisition And Overnight Memory Consolidation. Journal Of Research In Music Education, 62(1), 89-99.
Everyone was instructed to “play the melody as quickly, accurately, and evenly as possible,” but those in the listening group were also presented with a recording of the passage, which they heard 10 times before doing any practicing.
12 hours later…
Approximately 12 hours later, after a night of sleep, they returned to the lab to see how much of the previous evening’s practice stuck.
So how did the groups do?
Well, both groups improved with practice, in that everyone performed the passage more accurately after 12 minutes of practice than they did on their first try (no surprise there). In addition, both performed more accurately the following morning than they did at the end of their practice session. Yes, you read that correctly – they continued to improve overnight despite no further practice. Which might sound a little surprising at first, but was actually something the researchers expected to find, as continued motor learning after a period of sleep is a phenomenon that has been observed in previous research.
Different rates of learning
Though both groups improved during the practice session, they did not improve at the same rate. The students who heard the recording started out with an average accuracy score of 52.18 and over the course of the practice session improved to an average score of 99.96 (a 92% improvement). The group which was not given a recording to listen to started out at about the same level of accuracy – 48.71 – but only improved to an average score of 84.16 (a gain of only 73%).
Different levels of performance
Not only did the groups differ in terms of the rapidity of their learning, but the two groups achieved different levels of performance too. The group which heard the recording continued to improve overnight (again, despite not practicing any further), and ended with an average final accuracy score of 108.75 at the next morning’s test (an improvement of 9% over the previous evening’s performance). The no-listening group improved a wee bit too, but only by 4%, finishing with a final accuracy score of 87.24.
The results of the study seem pretty clear, but it may still be interesting to debate whether listening first is always the best approach in the long run.
I think having an auditory model can certainly accelerate our progress in the early going and help us get to a higher level of performance quicker. But does this lock us into a particular way of playing a piece too soon? And make it more difficult to develop our own interpretation of a piece?
What have you found in your own learning? Do you find it useful to listen to recordings before looking at the score? Or find it more valuable to start from the score with no auditory model?
A version of this post originally appeared at The Bulletproof Musician