2 ms·

I want to reinforce ThomPete's point. Most of learning a musical instrument is muscle memory. I really don't think flash cards / spaced repetition will help y

I want to reinforce ThomPete's point. Most of learning a musical instrument is muscle memory. I really don't think flash cards / spaced repetition will help you for learning an instrument. In fact, it seems to me that it could actually make it more difficult for you because you would start to associate notes with visual neural pathways rather than motor pathways. To learn any activity involving muscle memory there's really no substitute for hours of actual practice and quality instruction.

pkroll 12y ago
Actually, that brings up an interesting question: does muscle memory act like "regular" memory, with a curve of forgetting and with spaced repetition being a viable solution? If it did, then you could adapt at least some of these SR programs to, say, "perform this piece" or "do the following exercise" and learning an instrument might get considerably easier.
- gwern 12y ago
  I'm glad you asked! The answer seems to be, while there's little direct evidence on music, in general spacing does seem to help motor activity although the evidence is not as overwhelming as for declarative memory: http://www.gwern.net/Spaced%20repetition#motor-skills http://www.gwern.net/Spaced%20repetition#motor-skills
- coherence 12y ago
  There's a great deal to performance which is more intellectual than motor, and I'd be very surprised if there were no spacing effect for either aspect. My experience is that this kind of spaced repetition is dramatically effective. I'm working on an app (http://pianopracticeassistant.com/ http://pianopracticeassistant.com/) to manage it, since I found that Anki is a poor fit for the real structure of practice. I started it to help me learn a pretty huge piece (The People United Will Never Be Defeated) with very limited practice time. I've made way more progress in the last six months managing my spacing/interleaving this way than in the six months before that with equally sparing practice. Like gwern says, there's not a lot of direct evidence on music. I'm not aware of any studies in a laboratory setting that were equipped to directly show realistic spacing effects, as opposed to "non-musicians who played a fifteen-note passage once per day remembered it better." (Gwern links the Stambaugh 2009 study in his overview, which is a good one but which I interpret as more about interleaving than spacing.) But even though there aren't good lab studies, there are good expert case studies (The Practice of Practising has a good set [1]). Actual real-world effective practice, as done by experts, seems to combine long-term spacing of practice between sessions with massed focused work on subsections within practice sessions. You also see interleaving patterns of section-by-section and integrative practice ("work and runs") in both the long- and short-term. Amateurs on the other hand tend to practice mostly in runs (playing things all the way through), and I'm not sure how much spacing alone can help otherwise ineffective practice. [1]: The Practice of Practising, a figure showing "work and runs": http://books.google.com/books?id=IW9wKssw2hsC&lpg=PA9&ots=zgcqrs77Ha&lr&pg=PA14#v=onepage&q&f=false http://books.google.com/books?id=IW9wKssw2hsC&lpg=PA9&ots=zg...

pkroll 12y ago

Actually, that brings up an interesting question: does muscle memory act like "regular" memory, with a curve of forgetting and with spaced repetition being a viable solution? If it did, then you could adapt at least some of these SR programs to, say, "perform this piece" or "do the following exercise" and learning an instrument might get considerably easier.

gwern 12y ago
I'm glad you asked! The answer seems to be, while there's little direct evidence on music, in general spacing does seem to help motor activity although the evidence is not as overwhelming as for declarative memory: http://www.gwern.net/Spaced%20repetition#motor-skills http://www.gwern.net/Spaced%20repetition#motor-skills
coherence 12y ago
There's a great deal to performance which is more intellectual than motor, and I'd be very surprised if there were no spacing effect for either aspect. My experience is that this kind of spaced repetition is dramatically effective. I'm working on an app (http://pianopracticeassistant.com/ http://pianopracticeassistant.com/) to manage it, since I found that Anki is a poor fit for the real structure of practice. I started it to help me learn a pretty huge piece (The People United Will Never Be Defeated) with very limited practice time. I've made way more progress in the last six months managing my spacing/interleaving this way than in the six months before that with equally sparing practice. Like gwern says, there's not a lot of direct evidence on music. I'm not aware of any studies in a laboratory setting that were equipped to directly show realistic spacing effects, as opposed to "non-musicians who played a fifteen-note passage once per day remembered it better." (Gwern links the Stambaugh 2009 study in his overview, which is a good one but which I interpret as more about interleaving than spacing.) But even though there aren't good lab studies, there are good expert case studies (The Practice of Practising has a good set [1]). Actual real-world effective practice, as done by experts, seems to combine long-term spacing of practice between sessions with massed focused work on subsections within practice sessions. You also see interleaving patterns of section-by-section and integrative practice ("work and runs") in both the long- and short-term. Amateurs on the other hand tend to practice mostly in runs (playing things all the way through), and I'm not sure how much spacing alone can help otherwise ineffective practice. [1]: The Practice of Practising, a figure showing "work and runs": http://books.google.com/books?id=IW9wKssw2hsC&lpg=PA9&ots=zgcqrs77Ha&lr&pg=PA14#v=onepage&q&f=false http://books.google.com/books?id=IW9wKssw2hsC&lpg=PA9&ots=zg...