What does it mean to have learned a piece of music? Is it being able to play through it, start to finish, at tempo, without mistakes? If so, what does it mean if you try again a few days later and make a bunch of mistakes—have you un-learned it?
Musicians often struggle to build up and then maintain a repertoire of pieces or songs they can play. Doubly-so if we’re talking about playing them “off book”, without notation. It can feel like every time you put something new in, something old gets forgotten.
Another common complaint is that the more music you have in your repertoire, the more time you need to spend keeping it all at a playable level. Soon your entire practice time is taken up just “refreshing” the music you know you should be able to play, and there’s no time (or brain capacity!) left for adding anything new in.
We once had a Next Level client, for example, who was spending 2-3 hours per day on piano practice—but after scales and warmups, all that time was spent just playing through learned pieces. Why? Because otherwise she found she would “lose them” and have to pour endless hours into learning to play them all over again. As a result she had really struggled to add any new pieces or increase her skills. It took all the time and energy she had just to stay in the same spot This is one common cause of the dreaded “plateau”, where your ability level seems to stall out at a certain point and it’s hard to make any further progress. Another common cause is lacking the other Superlearning tools in this chapter, so that sticking points and challenging music seem insurmountable—because no amount of massed repetitions ever moves the needle for you. Fortunately that won’t be a problem for you any more! .
All these problems stem from a misunderstanding of the learning process and how the brain’s memory system works.
When it comes to music practice, we tend to use the words “learn” and “memorise” loosely, and to mean two different things. For most musicians, to “learn” a piece means practicing it to the point where they can reliably play it mistake-free, at tempo. To “memorise” it means they can do the same thing, without the visual context of sheet music or other notation in front of them.
What’s fascinating is that from a brain science perspective, all of that is, in a sense, “memorisation”. Even if we’re relying on notation, and even if we can still only half-play the piece correctly, our brain is going through a process of forming new memories and recalling them. In fact, the “mental representation” we’ve been referring to in this chapter, which lets your brain instruct your body on what movements to make when, is in reality a “memory” in the brain.
What is “Memorising”?
Just like scientific researchers have studied how so-called “superlearners” (a.k.a. “prodigies”, “the gifted”, “talented people”) can learn new skills ultra-fast, they have also studied how the top memory champions can do what they do.
The people who can memorise the order of a pack of playing cards in under 60 seconds, or memorise hundreds of random words or numbers in a matter of minutes (such as in the World Memory Championships) use a range of practical techniques to accomplish these seemingly-magical feats. We’ll cover the most relevant one for musicians below.
Aside from specific techniques though, there is one massive lesson for us from the world of memorisation research. Like all things Superlearning, it seems odd at first, but makes a lot of sense once you understand it.
If the central concept of Superlearning is Deliberate Practice, with Contextual Interference as a powerful toolkit to support it, then the second most important concept is to distinguish between Encoding and Retrieval.
The average musician thinks of “memorising music” as something you do to get off-book and be able to play without notation, and the method used is usually massed repetitions i.e. just try to keep cramming it in there, and hope it sticks.
There are two problems with that. Firstly, as noted above, memorisation is happening throughout the music-learning process, not just when we want to get away from the visual cues which notation provides. And secondly, massed repetition is once again the least efficient way to achieve our aim.
The research reveals that there are actually two distinct processes involved in memorisation:
- Encoding is the process of building up the brain’s “mental representation” of the music we’re trying to learn to play i.e. “getting the music in there”
- Retrieval is the act of accessing that stored representation, to play the music i.e. “bringing the music out again”
And here’s the key: both of these require dedicated practice effort! So one way to look at the traditional way of trying to memorise music is that we’re using a very inefficient Encoding process… and then not actually practicing Retrieval at all. We just try to brute-force the Encoding, and when it comes time to perform, we hope the Retrieval goes well. If it doesn’t, we assume that means we haven’t practiced enough, and need to spend more time “getting it in there”. In fact, we simply needed to spend more time practicing “bringing it out again”.
So the opportunity here is to recognise these are two distinct processes, and be intentional about when we are focused on Encoding new music, and when we have completed that stage and are ready to move on to practice Retrieving it.
We’ve already covered how to Encode new music more efficiently: Deliberate Practice, Contextual Interference, and most everything we’ve covered in the chapter so far has been about that process of “learning to play the music”. Now it’s time to examine what “Retrieval practice” might be, and how to make it as efficient and effective as possible.
Getting “Off Book”
We have just broadened our definition of “memorising the music” to encompass the whole process of learning to play it, whether with the visual cue of notation or without. It’s worth clarifying then that what we’ll cover in this section can be applied in both cases.
Even if you only ever perform the music with notation in front of you, the same distinction between Encoding and Retrieval applies, and the same technique of Retrieval Practice (presented below) can be used.
If you do want to get “off book”, and reliably and successfully bring the music out again without the visual cue of notation in front of you, that will most often be done as an additional phase of learning the music. Once you have successfully Encoded “how to play it”, and can reliably Retrieve “how to play it”, you can work on Encoding “how to play it without the notation” and practice Retrieving “how to play it without the notation”.
With that being said, some musicians prefer to incorporate getting “off book” from the outset, as part of the initial Encoding phase, and we recommend this. Or if you’re learning a piece by ear rather than from notation, that’s actually your only option.
Again, all the same principles apply. It’s just a choice of whether “getting off book” is part of what you include in the initial Encoding phase, or leave it as a later separate task of Encoding and Retrieval Practice.
Retrieval Practice
If our goal is to intentionally add a Retrieval Practice stage to our learning of a piece of music, the first step is to discern when the music has been successfully Encoded and is ready for retrieval. This isn’t a black-and-white question and it’s not always obvious. But roughly speaking, once you’ve ironed out the kinks and any sticking points, and you can play through a piece successfully, the piece is probably ready for you to begin Retrieval Practice with it. One good indicator is when mistakes occur more randomly, rather than it always being the same “problem spots” which trip you up.
In reality, it’s not a strict “A then B” process. There can be some back-and-forth between the Encoding and Retrieval stages, as Retrieval Practice reveals problem spots that actually need a bit more Deliberate Practice to firm up that mental representation. Or, contrariwise, going through Retrieval Practice may actually resolve some issues you thought would require a lot more Deliberate Practice in the Encoding stage. However, for the most part, our overall process is to get a new piece through Encoding and then into Retrieval Practice.
Note that pieces do not necessarily have to be up to target tempo to begin Retrieval Practice. We’ve found that pushing something into Retrieval Practice sooner than you think you should can, in fact, be quite powerful.
Retrieval Practice isn’t a complicated or mysterious process. We are literally talking about simply practicing “bringing the music out again”.
If you take just one idea away from this section, let it be this: when you return to a piece of music and struggle to play it, that may not mean you haven’t learned it well enough. It probably just means you need to practice that act of retrieving what you previously learned.
The Power of Forgetting
Let’s come back again to the traditional method of learning or memorising music. We spend all our time “cramming it in there”. Then some time passes, and then we try to “bring it out again”.
What happens in the middle? There is some level of forgetting. We think this is a bad thing, so if we want to “learn it better” or “memorise it better” we generally try to do more “cramming it in there”.
In reality, just like mistakes can be our friends once we understand Deliberate Practice, that forgetting stage can be our friend, once we understand Retrieval Practice.
The reason we struggle to bring the music out again is that we aren’t giving ourselves enough opportunities to practice that middle step: time passing, before we try playing it again. Doing more “cramming it in there” only makes that worse!
Retrieval Practice is about turning forgetting to our advantage. Similar to how Contextual Interference deliberately introduces additional challenges so that we can learn to overcome them, Spaced Repetition is the idea of deliberately allowing ourselves to “forget” a little, so that we can learn to overcome that.
Spaced Repetitions
Spaced Repetition means we intentionally design how much time passes between our attempts to Retrieve a memory. The goal is to wait just long enough that it’s a bit of a struggle to bring that “mental representation” up again. If we play the music again too soon, we aren’t practicing Retrieval. If we wait too long, the mental representation (i.e. the memory) will really fade, and we’ll need to go back to Encoding it again.
It’s important to note another similarity with Contextual Interference: this can be brutal! We are intentionally setting ourselves up to struggle. As long as you know that, you can be prepared and not let it discourage you. Don’t be surprised if you find yourself making a lot of mistakes and feeling like “things are getting worse” when you first try using Spaced Repetitions. Have faith that the music is well encoded, and that forgetting will lead you to remembering.
Spaced Repetitions are the bread-and-butter of Retrieval Practice. We play the piece once and only once, with no “do-overs” or corrections. Then we wait for a certain interval of time before repeating that.
We leverage the way the brain gradually forgets by increasing the interval each time. For example, you may begin by interleaving your repetitions in between 10 or 15 minutes worth of other practice, maybe three times in a day. Then perhaps we reduce it to once a day, then once every two days, and so on.
Just like in Interleaved Practice (covered below) and the “80% correct” guideline for Contextual Interference, Retrieval Practice recruits more of our brain to the task, by introducing an open loop e.g. “I retrieved that piece and it was awful. While I’m sleeping, brain please go ahead and figure out what needs to be adjusted or reinforced in my mental representation to bring this out when and how I want it.”
Spaced Repetitions take time to work their “magic”. When embarking on a course of Retrieval Practice, it’s good to give yourself 4-7 days before you evaluate the results. After that time, you can go through the music several times at different metronome speeds and see what happens. If needed, you’ll have plenty of other Superlearning tools to then make any other improvements necessary.


