Most of the words of wisdom that I've read about embedding a piece in your psyche say that each "unit" (bar, phrase, section) needs to be repeated correctly X times. There is a lot of discrepancy over how many times though... 3, 10, 50, 100? Where do these figures come from? Is there any research to back up the figures? How many times do you repeat a unit, or what do you tell your pupils?




Please only vote in the category which *most* applies to you, even if all three categories apply in some way (although you still have to tick the box "I've voted elsewhere" in the other two categories, otherwise the system won't accept your vote)

Great question skylark - one I am struggling with right now - having done something like 50 repetitions each day for the last 6 weeks and still cant rely on getting it right.

Couldn't get the voting to work though

Hmm. I repeat things as needed and work on them until they are consistently not wrong (i.e. until they are reliably correct rather than just repeating until I play it correctly). This might be once, it might be thousands of times over numerous practice sessions - it depends on what is needed.

Not sure if that helps in terms of what you're asking on this thread, but it's the only answer I can honestly give.

I must admit I don't keep an accurate count, but I'm trying to improve a phrase/section at the moment and I've repeated it until I can play it smoothly step by step up the metronome, but then I get to a certain point (which is well under the tempo it should be played at) and I can't get past it

You have to vote in each of the three categories

[quote]Couldn't get the voting to work though [/quote]You have to vote in each of the three categories [/quote]

Didn't read instruction - will try again!

Actually I think the instruction is very unclear - my fault! I'll go and re-word it

...and it worked!!!

I know what you mean, and it might be difficult to put a figure on it. I think what I'm asking is that although we might *think* it's reliable after, say, 20 correct repetitions, is there research which proves that it isn't truly reliable until you've done 30... or 50... or 100 Otherwise the figures seem to be plucked out of thin air, depending on what book you're reading...

I think it's quite important when you do the repetitions. I've read somewhere that ideally you would do something 5 or 10 times (depends how long the something is), then leave it for 10 minutes and repeat. Repeat again after 1 hour. Repeat again after 4 hours. And so on.

I certainly find that I learn things more quickly by doing them at different times than by hammering at them in one session. For instance, if I have a particularly tricky bar, I might keep playing it 5 times over at intervals throughout the day. What's annoying is that I sometimes find that the next day I'm right back to square 1! But on other occasions, leaving it overnight seems to work wonders and it's no more trouble. This could be something to do with the total number of repetitions, as you suggest, but I really don't know - it's just that at some point it will click and you'll be able to play it ever afterwards.

(Haven't voted as I'm not sure how many times is right!)

Interesting thread and interesting question. Being an adult, I am so concerned about how long it takes to learn anything that I often need telling that things take veeeeery loooong to learn, then I feel comforted and that gives me the boost to carry on.

This is most certainly true - and is a good reason for why short frequent lessons are more effective for a beginner than longer more infrequent lessons. But once we're sure (pupil or professional) that we're barking up the right tree with regard to what we're aiming to make fluent, with all the right articulation, etc, as well, I don't think there's any way round repetition. I haven't actually voted, by the way, as it one of those 'it depends' issues! It depends how hard it is!

Yes. If it's not sight reading, or something that we just want to 'get through' with rhythmic accuracy being more important than perfection, as in accompanying, then 'practice' means making sure we get it right before we aim for fluency. Cemented-in mistakes are harder to fix than if we were starting from scratch.

Flossie is right. There is no answer. The variation is huge. From a single repetition, to never mastering it no matter how many times you repeat it

It depends on:

Whether you are familiar with other pieces in the same style
Whether you are already familiar with the movement patterns
The difficulty of the piece compared to your general level
How much preparation you have done away from your instrument
How well rested and alert you are
Your state of health
How important it is to you
The intensity of your concentration
The accuracy of your initial attempts

... probably many other factors

Erm, as many times as are required, whether that's once or hundreds of times!

I wonder if it's possible to over-repeat a passage, to the extent that you get past the peak performance at that tempo and start going backwards again...

I've been playing a section really slowly, and thought I was never going to move up a notch. I'd been playing it mostly quite well but then instead of improving further, I started making more mistakes again. I wondered if I'd lost concentration through constant repetition at the same speed, so instead of putting the tempo back down, I increased it slightly. Eureka! I found I was actually playing it better. My hands discovered a momentum and seemed to find their own way to the keys. A chord which I'd been struggling with, suddenly came right because my hand position automatically adjusted itself so that the fingers fell naturally into the right place.

Am I deluding myself that going up a notch in tempo was a good thing to do? Is it possible to play too slowly or to repeat something too often? Thoughts anyone?

This is really subjective to be honest. Something simple like a typical Einaudi piece might click straight away, while something with a few voices (eg: many of Schumann's works) may take several, if not more attempts to get it ingrained into memory.

Example:

Einaudi's Indaco took me just under a week to get it about right and it's 6 pages.
8 bars of Schumann's Traumerei took me about a week to memorise properly.

I also think it is subjective but think no amount of repetition - if properly done, of course - will spoil learning. If learning is spoiled, I would suspect there is something wrong in my repetitions.
If repetition is properly conducted (possibly mixed with variety, e.g. slow and even extra-slow tempo, different rhythmic patterns, doubling of notes in tricky passages, etc.) it should lead to assimilation and long-lasting learning. I very often find that playing passages very very slowly brings out mistakes that are sort of lurking and waiting to surface when one is under stress (exam, recital, playing for somebody else, etc.).

Definitely not. A lot of time the only way I can achieve fluency is to speed it up! So you are not alone.

Of late I find that different pieces need different approaches to learn. On your piece in question could slow the whole piece down to the same speed as the tricky part, then slowly but surely speed it up over a period of time. I find this naturally happens with many of my pieces until I get the metronome out and discover I am playing too fast! Then I have to slow down to the correct speed. I don't necessarily approach all of my pieces like this but this is just one particular method I use.

I think that if you've mastered a passage, then you should be able to play it doubled or halved speed without it being less secured at all. (In fact, if I intend to play something at, say, crochet = 60, I'd make sure I can play it as securedly at, say, 80 and 40 as well.) Sometimes adjusting the speed can help, say, with shaping a phrase, but not fluency, I'm afraid.

Also, I don't believe in repetition for repetition's sake. Obviously, practise is repeating, but I don't think it means simply repeating what you see on the page in front of you over and over again until you "get" it. In many cases you need to think how you should practise something or what the problem is that's hindering you. A very simple example is that a running passage might turn out to be a broken chord in disguise (with a different note order to how we'd normally practise a broken chord/arpeggio so you don't spot it straight away) and so should be practised as chords as well as the usual different rhythms etc.

As for the issue of over-practising, MT will come along to recommend you sleep over things and let the subconscious do the work!

Even if they can't correctly recite the entire journey, do you not find they remember a few more stations after the second and third repetitions? I'd be surprised if there was no improvement in recall. And would this process have a different result if using data (i.e. words) with which they were already familiar (rather than stations some of them will never have heard of); even something simple such as colours, for example?

It's an interesting idea and I agree that there is a lot more to practice than repetition, but I'm not sure that your demonstration is conclusive or proves that repetition is ineffective. A novice won't be able to learn a challenging piece after one playing, so surely some repetition, along with other strategies, is necessary?

A couple more of perfect ones beyond the first time it's right. And one less than the one which introduces a new error.


Repeat in subsequent practice sessions as needed .

For better feedback, can you tell us what the section that you're trying to improve is?
Bye for now,
JC

Now ... this is really good advice. And so terse and precise.

Yes, I thought so too! You should write a recipe book, anacrusis, on practice, with ingredients, method and cooking time, depending on the oven you have. Hey, that's actually a good idea... I wonder if it'd sell well...

The advice above and on this thread is very interesting, and clearly made me realise as a beginner it is not always obvious how to practice effectively. I hope this is something I can learn as time passes.

...and I thought I was practicing effectively already.... WRONG!

I think you'll find that some people might be able to recall all or most of it, using association techniques, which is a completely different process to memorising music, I'd wager I went on a course years ago and was shocked at how much it was possible to remember using special techniques

Ok, then Solari, share!

One thing that helps me memorise a bit is visual association. I cottoned onto this when the phone rang twice in quick succession, each time when I was three bars into the second page of the same piece. So after that, every time I got to those bars a vision of the person who had rung popped into my head (it was the same person each time) and I sort of thought, "Oh yes, this bit." So after that I started telling myself little stories when I was playing and it helped me link up bits that had been isolated sections up to that point. But I'm still not good at memorising. What I'm talking about is muscle memory, really. As in, to put it crudely, that hand goes there when this picture pops into my head, which it does at that previous point.

In a similar vein, we used visual association, and also the fabrication of a story. We were tasked with remembering about 30 items on a shopping list and managed the lot by creating a story around each of the items.

Another method was to imagine moving from room to room, associating objects with each room. It worked well. Shame I can't figure out anything similar for playing the piano and mechanical memory!

This is interesting. I had never thought of doing it this way round. I learnt things for my A levels while playing the piano and in the exam I would think of what I had been playing and recall whatever fact was associated with it.

I've always been taught that the correct approach to learning a new passage of music, is to play it very slowly, eight times.

Works for me

butbut, I'm a picky eater, and I don't do scales, not even deep fried. I have to chop them up small and present them in a sauce so I can't pick them out too easily .

Scales are like vegetables - they don't suit everybody, and not everybody needs them. We can get the nutrition from other sources like fruit - or music in this case.

Sorry to transgress....

I too discovered this, completely by accident. I'd been hacking through a difficult part whilst watching Countryfile (probably not to be recommended), thinking I hadn't been paying much attention to the TV. When I came to play said part with my chamber group, an image of pig arks came into my head each time we played a certain passage! Every time I play this piece now, I still think pigs. But rather than being a distraction I have found it helps me to focus because even if I'm struggling with other parts of the piece or my concentration is waning, I know the "pig" section is coming up and that I have mastered it. Weird or wot....

As to how many repetitions are needed, I take a similar approach to Anacrusis. Particularly if a passage is difficult, it is not adequate to just play it right once, or three times, or whatever, and then move on. You have to cement the right way of playing it into your mind and your muscles. Depending on what it is (evil 3rds scale, I'm talking about you again), there are times when I doubt 100 repetitions would convince me I had mastered it. Better plough on then!

Actually I think the simple answer is yes.

But the underlying psychology is more complicated. If you are practicing correctly - with full concentration, then additional repetitions ought only to strengthen your grasp of the piece. That they sometimes do not is I reckon because once you can play something OK (but not yet as well as you'd really like) there is a danger of repeating it over and over relying purely on kinaesthetic (aka muscle) memory, even though you may have the score in front of you - by the time you know the piece you are probably not following it carefully, but just using it to cue each section, and maybe guide you through one or two especially tricky bits. Then when something goes wrong (you forget a bit, play a different piano with different touch sensations, play when your mood is a lilttle different) you have no back-up of intellectual understanding, and your sense of which tones map to which keys is too weak to wing it based on auditory memory. So a little bit of the piece disintegrates.

It is similar to falling apart during a performance of something you play well in practice. The cure is four-pronged (possibly more-pronged):

1. Continue to study the score - structure, motifs, harmonies, shapes, landmarks, ...
2. Now and again vary your practice location and time
3. Do even more slow, thoughtful repetitions to burn the piece ever more firmly into the memory
4. Develop confidence and trust in your recall abilities


I have never been a fan of getting up to speed by ramping up the metronome a notch or two at a time. It creates stress, leads to sloppy and inaccurate playing, and it does not achieve its goal well either.

I have found it more effective to continue to work on the piece at a far slower tempo than I can actually manage. This eventually leads to a high level of precision in the movements, relaxation (very important), and steady pulse, alongside very strong memory of where the hands and fingers are going. It is then usually easy to just play the passage at any desired tempo. When you really know a piece you will usually find that you have the opposite problem - holding back the speed to what now seems rather slow.

It is similar in sports. You don't learn to run/swim/cycle faster by constantly pushing yourself to go flat out. That leads to injury and burnout. The biggest improvements come from sorting out your technique - and you practice the improvements at a comfortable speed before seeing if they hold together and give you an improvement when you are really pushing it.

It is interesting that when you know a piece it seems that time expands to encompass it as you play it. The same is true for listening. When you first hear a new piece it can sound very fast, and you may not even hear all the notes separately. By the time you know it well you hear everything that is happening and you'd swear that it was all happening much slower. But it is the KNOWING it that makes the difference - not trying to force anything.
Sometimes there is some small element of technique, or just one or two awkward points in a passage that hamper attempts to play faster. You then have to find them, find the correct technique, and practice it until it is a part of your equipment.

But deliberately holding back on speed in this way takes a lot more discipline and patience than the "lets ramp it up to speed" approach.

I was always told that trying to learn phrase (or whatever) by repetition while making mistakes, means the mistakes themselves become learned, and that's hard to undo, so the phrase has to be note perfect before trying to learn by repetition, so that what the brain remembered is note-perfect. And then I seem to remember that a miminum of three repeats was required. I'm afraid I have never been patient enough to put it into practice, so I don't know if it's right.

It always seems to take me ageeees to learn a new piece purely by repeating it over and over I have a memory like a sieve though what seems to work better for me is learning, say, four bars, perfectly, then another four bars, then play those eight bars through, working on any bits that need more attention, then carrying on like that until the piece is learnt

I'm just catching up on this thread as I've been away for a few days. I've discovered that my pieces are better for having had a break from them, but there are still one or two bars which are insecure so I'll try some different approaches to see what the result is. The responses have been very thought-provoking - thanks for the contributions.

In other words, practicing properly is more effective than practicing incorrectly. Repeating an entire piece over and over is just about the worst possible way to try to learn it

You do short sections to overcome technical difficulties, and when memorizing.

Apart from initial explorations to get an idea of the piece, you do complete play-throughs when all or most of the technical difficulties are solved and you are trying to achieve fluency and overall balance.

On the oboe (and probably other wind instruments) it is also a good idea to do complete play-throughs for time to time to develop stamina.

I learn a new bar very quickly by analysing it. What chords are represented and in what inversions? - and spot any passing/non-harmony notes and whether they're on the beat? That generally slips it into the brain straight away.

How many times do I repeat a unit?

Do you mean in a single practice session?

Rarely more than a dozen times, usually five or six. The improvement usually becomes apparent a day or two later, and experience shows that repeating more times than that does not result in much (if any) extra improvement.


Or do you mean in the course of learning a piece?

In that case it might be hundreds of times over the course of weeks or months, maybe even thousands.

My flute teacher taught me a technique for overcoming errors that my piano teacher never once mentioned, so I don't know if it applies readily to the piano or not.

If I meet a phrase which I am finding difficult to play even after a few weeks of learning, she makes me play it backwards. Then she makes me play it forwards but with a different rhythm from that written, ie if it is quavers, then dot a quaver and follow it with a semiquaver, then do the same for the next pair. Then reverse the one with the dot and the semiquaver. Then do all of this backwards. Then play it forwards grouping every three notes into triplets. Then backwards again as triplets. Naturally all these variations are much slower than the desired tempo, but when I come to play it properly forwards again, it seems to be much easier to play.

Naturally, this process is quite difficult to do backwards if the phrase contains accidentals as you have to look ahead to the left of the bar to see if you have to play any, but maybe this is also part of the increased concentration which then leads to the same level of concentration when playing it normally. I don't know how it works, but it has been helpful to me in a number of pieces.

Now, sightreading backwards, that would be a challenge!!

You know I don't think I've ever truly tried to play a piece backwards. Difficult on the piano I would assume, due to the plethora of notes and of course encountering at the far end of the bar a note that ought to have been held for its entire length!

With what piece do you think I should take up this challenge?



Sightreading blindfold would be something else again!

My clarinet teacher uses the backwards technique too.