Amount of practice - procedural knowledge

Hi

I recently have been getting ready to play jazz guitar in groups, so I've been using a program called 'guitar scales method' to practice. I already play jazz bass in a group but my guitar skills needed (and still need) quite a bit of work so I've been getting up to speed and using this program as an aid.

Anyway, I was surprised that for some of the exercises, the guy who wrote the program specified to not do more than 5 minutes per day, and to not do more than one variation per day, but to rotate them. So, at the early stages there are about 4 exercises you can do, but only for about 5 minutes each per day. All four of them are what I'd call 'brain' exercises since they involve finding notes, positions, etc. It's not just running up and down scales, which he doesn't put a limit on.

So being impatient for progress, I decided to do some research to find out what the logic behind this was, and I found out some interesting things. I didn't go to the guy who wrote the program I looked in journals.

First of all, in psychology, the type of knowledge that playing an instrument involves (if it's put into practice, not just theory) is called 'procedural knowledge'. Procedural as opposed to 'declarative' knowledge (knowledge of facts) or 'personal' knowledge (knowledge of biographical events). The key thing with procedural knowledge is that it is task-related. For example, knowing that the major scale is TTSTTTS is one thing, playing it is another. Playing it is procedural knowledge.

I had a look through some recent scientific papers which I could reference in case anyone wants to look them up - (I don't have the URLs, but I have printed them so I have the titles and authors) - to find out how procedural learning works, and found out some interesting things.

To put it succinctly,

1. Most goal based procedural learning happens as you sleep. The 'goal based' part means the actual thing you're specifically aiming for, rather than the motor skills of getting to it. Motor skills or physical action can be improved during the same day. For example, finding different notes on a fretboard by note name is goal based. Playing lots of different notes without seeking them is just physical. This seeking activity improves mostly at night as you sleep (it seems like the brain rehearses it along with other things you did that day).

2. After a certain number of repetitions, there is no added advantage to continuing a goal based exercise on the same day. The number of repetitions was quite limited I seem to remember.

3. When several very similar goal based tasks were practiced on the same day, the learning of the original exercise was reduced by the learning of subsequent exercises.

What this means is that the best way to learn is to divide up (A) what you actually have to think to do (finding chords or notes quickly)and (B) what is purely physical (running up and down scales, finger strength etc).

A - each task related to this should be done for about 5 minutes per day, on rotation if for similar things. For example if you are learning 5 positions of a scale, do one for 5 minutes each day, Monday-Friday. Anything more than that per day is a wasted effort. The key thing with this type of knowledge is doing it every single day over a long period.

B - you can work on motor skills as much as you want and you'll show improvement, as long as you don't get tendonitis etc.


Hope people find this useful - I would recommend doing your own research in this area if you want because there's lots being found out at the moment about how the brain works.

[Billnc]

Cool! makes sense, so I'm presently doing a little exercize where I play through a Parker tune and use only chord tones, but not chasing roots etc. I may add an approach method etc. Is this procedural? I do stuff where I have to THINK hard and play slowly. I take it also that the tunes should be broken down to four measure chunks too? Is that better than the whole tune?

Now just walking through them or learning the heads is unlimited right? Same with the actual arpeggios.

[TheFrogPrince]

Bill, playing through an old tune with associated arpeggio notes is basically just a finger exercise and maybe a memory aid. It's akin to going over your notes from American history back in said grade where you took said course for your exams. You know that the Constitution was signed on August 1st, 1776 somewhere in the subconcious but are just refreshing it.

Where if you were learning a new tune, it's a new skill to acquire as you can't play it through smoothly with few or no mistakes yet. If you have to focus and go slowly, then it's a new skill. To use another example, learning how to balance molecular bonds in chemisty using atomic charges and all other stuff (i.e. two molecules of hydrogen attracts two molecules of oxygen forming hydrogen peroxide etc etc).

[Billnc]

Thanks! Now, can you break it up, Like learn a song for 5 minutes, next session of the day a new scale or is it one thinking thing a day?

Hey

What i've been using this system for is not songs, so I'm not sure how it would be put into practice for songs.

But for theory, if you were to spend 5 minutes intensively finding notes on a certain string, that would be one exercise. One string per day for this exercise would be the way to go.

But if an additional exercise is different enough, you can do that too. So for example if you had an exercise where you practiced playing random intervals, you could do that also for 5 minutes and it wouldn't obscure the original learning.

Another exercise could be learning digital patterns in scales. So for example playing ascending intervals like 3rds, 4ths, 5ths, 6ths, in a scales. While this is new (so you actually have to use your brain to do it) then 5 minutes per day would be enough.

You then could end up with 4-6 very intense exercises all for only 5 minutes, adding up to 30 minutes of very intense theory practice. If there are 5 positions of a given scale, you could rotate the same exercises from monday to friday using 5 positions.

For the stuff you already know, like familiar scales and interval patterns and songs etc, you could do as much of that as you like.

[bassandbeyond]

This is very interesting.....

I'm a big advocate of breaking new material down into small chunks to be mastered one by one, but only 5 minutes per topic per day??? That might be adequate for learning some procedural tasks, but I can't think of any great musicians who say they only practiced 30 minutes a day in their formative years.

How about 5 minutes at a time instead? In my own experience, I seem to progress most efficiently when switching between exercises every 10-15 minutes. But I might cover a given topic several times in the same day.

[xzzy]

Originally Posted by bassandbeyond

but I can't think of any great musicians who say they only practiced 30 minutes a day in their formative years.

But that's not really what the OP is saying. Nowhere did he say you can only practice 30 minutes a day. He just suggests limiting repetitions for a specific brain-intensive task to 5 minute blocks. This minimizes the effort you have to invest to convert the knowledge into "muscle memory".

Once you finish your exercises, then you can move on with writing songs or jamming with some friends.. that is, repeat the stuff you already know to work on your motor skills.

I think we all instinctively know the value of "sleeping on it", every single one of us has had a problem bugging us as we go to bed, only to wake up the next morning with the perfect solution in mind. This learning system is merely maximizing our use of that talent.

What I'd be curious to know is if the time of day when doing your exercises has an effect.. that is, if you do your 30 minutes when you get up in the morning, is there more or less benefit to doing it just before going to bed?

[bassandbeyond]

Yes, I understand that this system allows for unlimited repetitious practice of already-familiar material. But, if I understood the OP correctly, new material is supposed to be limited to no more than 5 minutes per day, per subject, and the number of new subjects tackled per day is also supposed to be quite limited.

What I'm saying is that, IMHO and IME, more significant progress can be made by revisiting new material multiple times per day in brief time increments. I also feel that spending lots of time on already-familiar material is of limited benefit.

Obviously, since we are not all robots, each of us is likely to experience individual variation in ideal practice methods. For me, the above-described routine seems to work best. And I have definitely found that I get best results when practicing first thing in the morning, when my mind is clear and I can concentrate best. Yes, I am one of those rare morning-people!

[TheFrogPrince]

Originally Posted by bassandbeyond

Yes, I understand that this system allows for unlimited repetitious practice of already-familiar material. But, if I understood the OP correctly, new material is supposed to be limited to no more than 5 minutes per day, per subject, and the number of new subjects tackled per day is also supposed to be quite limited.

What I'm saying is that, IMHO and IME, more significant progress can be made by revisiting new material multiple times per day in brief time increments. I also feel that spending lots of time on already-familiar material is of limited benefit.

Obviously, since we are not all robots, each of us is likely to experience individual variation in ideal practice methods. For me, the above-described routine seems to work best. And I have definitely found that I get best results when practicing first thing in the morning, when my mind is clear and I can concentrate best. Yes, I am one of those rare morning-people!

Nowhere did the OP suggest only practicing for five minutes with new skills. The author of the method book OP is using suggested practicing a given exercise from their material for said time.

What the OP said is practicing something tends to lose it's effectiveness after a certain number of repetitions whose number is apparently quite low. I can agree with that from experience.

I tend to be a perfectionist, never satisfied with what I play only to wake up the next day and be able to play with I couldn't the previous day. I play until my head hurts sometimes, a habit I need to break out of as it just burns me out quickly. Ergo, there is no real point to going over something for hours a day. It goes from learning a new skill to rote memorization and hand-strengthing exercises.

Originally Posted by xzzy

But that's not really what the OP is saying. Nowhere did he say you can only practice 30 minutes a day. He just suggests limiting repetitions for a specific brain-intensive task to 5 minute blocks. This minimizes the effort you have to invest to convert the knowledge into "muscle memory".

Hi

The 30 minute thing was just giving an example of how you can add together different exercises which all last 5 minutes to make a longer routine that visits many different areas. Extending this you could make an hour routine etc.

Also, the idea isn't to convert knowledge into pure 'muscle memory'. For example, one thing that you could practice is instantly being able to name any given scale degree for any given key. This involves no muscles, but is 'goal-oriented'.

I was not suggesting that total amount of practice should be limited to a certain amount of time.

Also in answer to an earlier question about time of day, the papers I read said that the more that takes place between the task being learned and the start of sleep, the less well that task is improved by the sleep. So yes, probably you would learn things better that you study at night than those you study in the morning.

I am no expert on this subject and just wanted to share some interesting ideas. If anybody would like to read a pdf of one of the papers I read I can email it to them. Actually I'll post it on my website it's probably easier

Here you go:
http://www.basslessonsbristol.co.uk/...me/lesson-pdfs

It's in alphabetical order and it's called 'PNAS-2005-Cohen-18237-41.pdf'.

Another one, which is a study on the role of repetitions

Ofen-Noy_et_al2003.pdf

again, at http://www.basslessonsbristol.co.uk/...me/lesson-pdfs

[bobknowsbass]

Interesting stuff, subscribed!

[bassandbeyond]

Originally Posted by afromoose

Hi
The 30 minute thing was just giving an example of how you can add together different exercises which all last 5 minutes to make a longer routine that visits many different areas. Extending this you could make an hour routine etc.
I was not suggesting that total amount of practice should be limited to a certain amount of time.

Oh, in that case I agree completely!

[Fergie Fulton]

Great post and a fantastic subject on so many levels.
In my own case i developed the skill to learn songs in the short term for performance, be this reading, listening or just being told.
With me if you aked me how the song went i could not tell you but once i start playing it comes to me. I have described this as like the start of Star Wars when the writing appears at the start. You can see it but cannot read it, but you can when it comes closer and into focus.
If i was doing a dep or a one off show where notes or sheet are not allowed, this is how i do it. Ask me a couple of weeks later to play the same songs and they are not there in my head to play, so i can't. But then promp me and i can. I would assume this is because of the way i learnt the songs, so many in a short period of time that they are there for short term recall with the correct prompt, but not learned as such to stay. By the correct promp i mean something as we go along, it seems familiar and a chord, vocal, drum fill, prompts my next action. I and others in jam sessions have similar experiences on songs we don't even know, when we change or stop for no reason other that "it felt right" it just happens.
Maybe that's a part of this proccess on another level as no real time was spend actually learning the songs because there were so many and not enough time.

[Billnc]

Thanks all for clearing up what I misunderstood.