Sunday, June 17, 2007

Motor Learning for the Musician: Making Practice Efficient

(The following article was in the Feb. issue of the banjo e-zine "Silver Strings")

Motor Learning for the Musician: Making Practice Efficient
by Kelly J Cole, PhD - Associate Professor
(Department of Integrative Physiology - The University of Iowa)

Many of us would love to emulate our banjo heroes who, during their 'woodshedding' days, spent hours playing the banjo each day. Our heroes exploited the first and most important principle of maximizing motor learning; more practice leads to more learning. Every other condition of practice runs a distant second in its effects on learning compared to this Law of Practice.

Let's face it, once you have the sounds in your head, those sounds, or something close to them, eventually will spill out of your banjo if you can practice hour-upon-hour and day-after-day. Those of us who have chosen not to give banjo playing the highest priority in our lives (in my case, this choice makes my wife and children very happy) have to budget our practice time. We're faced with the task of designing the most efficient practice that we can - a practice that produces the most motor learning possible per unit time.

There has been considerable research over the past century on just what practice conditions generate the most motor learning. Some of these principles can be applied to learning how to play a musical instrument. While some of these principles are counter-intuitive, none are controversial and appear in just about every textbook on motor learning published today.

"Motor learning" has been defined as a process that results in a relatively permanent increase in motor performance from practice or experience.

However, your performance at the end of a practice session is not a reliable indicator of how much you have learned. A better estimate of motor learning is how much you retain when you play well after the practice session is completed (hours later, or the next day). This is because temporary effects during practice sessions can improve or degrade performance (like changes in attention and fatigue).

Before going any further we must dispose of the term 'muscle memory.' Let's be clear. Motor learning results from experience-driven changes in vast networks of neurons in our central nervous system, not from changes in muscle. Muscle is best viewed as a motor and not as a structure where motor skill is somehow stored. Skill emerges from the precise and controlled activation of muscle, and this activation requires the nervous system.

Functional and structural changes begin in neuronal networks within minutes of practice. These changes continue during practice, and for hours and even days later as memory appears to consolidate. With this consolidation there are shifts in the brain regions that are most active, with the general shift occurring from regions more in the front of our brain to regions more at the back.

Muscle doesn't change as a result of practice, apart from changes in strength if practice involves the production of large forces, or endurance if practice involves low force and high numbers of repetition. So let's do everyone a favor and simply forget the term 'muscle memory' when talking about how we acquire motor skills.

The conditions of practice that best produce motor learning depend on the type of skill that we're attempting to learn, and on the stage of the learner who is attempting to acquire the skill.

Motor skills can be divided into those that occur continuously (e.g., driving a car, skipping rope) and skills that are discrete (e.g., throwing or kicking a ball, reaching accurately to grasp an object). The practice conditions discussed in this brief article assume continuous motor skills, which seems to fit playing a musical instrument, and particularly bluegrass banjo.

The best practice strategies also depend on whether we are just beginning to learn a new skill or if we have the basic idea of how to perform the skill but need improve speed, accuracy, reduce error rates, etc. Someone who is brand new to a task will benefit the most from frequent guidance (physical and verbal) and immediate feedback from an instructor, and from repeating the action without interruption a large number of times. This initial learning stage is meant to rapidly shape the performance, give the performer the 'feel' of the task, and provides the performer with a basic frame of reference for evaluating their attempts at the task without external guidance.

An experienced ballplayer who has never seen a curveball is considered a beginner at hitting curveballs and likewise would benefit from frequent guidance and feedback, and many uninterrupted repetitions. The experienced picker who is trying to learn a new and strange lick - like the player well-versed in the basic Scruggs style who decides it is time to learn to play 'Foggy Mountain Special' - also would fall into the category of a new learner. This picker, and our imagined baseball hitter, would quickly move from beginner to novice after perhaps only a single, intense practice session under these conditions.

Once you are 'getting the hang' of a skill, and are no longer a rank beginner, the conditions of practice that will produce the most learning change drastically. At this next stage of learning the frequency of feedback from an instructor should be reduced so that the learner can build a strong internal frame of reference for monitoring the correctness of their performance, whether it is the ability to hear when you've emphasized the proper notes in Sally Goodin' or the ability to feel when your right hand position is giving you the right angle of attack on the strings.

Similarly, the timing of the feedback can affect learning. At this stage of learning feedback provided after a delay, rather than immediately, has been demonstrated to have stronger learning effects as does feedback that summarizes several attempts at the task. This is because it forces the learner to rely upon their own evaluation and correct themselves. These issues concerning feedback from instructors, though interesting, probably aren't as important as how we practice when we're alone, which we'll consider now.

When we're new at a task we benefit most from lots of repetition of the same task, without interruption. But, as noted above, with this approach you will move quickly from a 'rank beginner' to someone who is getting the basic feel of the task. Intuitively, one would think that continued learning would occur at the fastest rate by maximizing the 'law of practice' and continuing to repetitively practice the skill without interruption. In fact, once you have the basic feel of the task, and you know when your performance does or doesn't have errors, this intense and repetitive practice schedule is not the most efficient way to learn.

There has been a considerable amount of research showing that we learn continuous motor skills much better when we take frequent, short rests. This type of practice is called 'distributed' practice, while the relentless practice of the same skill without interruption is called 'massed' practice. The problem with distributed practice is that it takes longer to achieve the same number of repetitions (don't forget the 'Law of Practice' - more practice, more learning). The key is to keep the rest period short, and refocus completely on something else, whether it is getting up for a drink of water or listening to a recorded version of a pro playing whatever it is that you are trying to learn. It seems quite likely that the short rest period helps you learn by keeping you from getting into a performance 'groove.' You want to practice all the internal processes required to perform the skill well on the first try - which pretty much defines a successful musical performance.

Repetitive practice of the same skill doesn't force you to practice everything needed to call-up and execute the skill the first time. A short break during which you refocus your attention on something else makes you practice the entire process when you return to the skill. So, after running through a difficult lick, phrase, or passage a few times you could take a break or play something else which appears to have the same 'refocusing' effect, as we'll discuss next. Let's say you have identified three things that you want to learn - a high break on 'Blackjack', the low part for 'Sledd Ridin', and fancy backup vamps that use slides to switch between the F and D positions of a chord. (Note: you've already 'got the hang' of each of these three skills). After warming up you could allot 15 minutes of solid practice (with short breaks) for each skill. This type of practice has been called 'blocked' practice because you practice a skill without interspersing practice of another skill. Conversely, you could jump from a repetition of Blackjack, to vamps for a minute, to a repetition of Sledd Ridin, and so on, in random order.

Research has shown repeatedly that you will perform the skills better during the practice session with a blocked practice schedule compared to practicing in a random order, but you will retain (that is, learn) more (tested hours later or the next day) when you use the 'random' practice schedule. One thought is that random practice keeps you from 'getting into a groove', and forces you to practice all the processes needed to recall and execute the skill correctly the first time. So, if you want to maximize the number of repetitions (adhering to the Law of Practice) and avoid the pitfalls of massed practice, the best approach might be to organize your practice around mini-blocks of three or so skills organized in random order. For example, play the skill three times (a mini-block) then switch to another skill for three repetitions, and so on.

Invariably, you will find a few spots in a skill that trip you up more than others. A consistent rough spot implies that you don't 'have the hang of' that little part of the skill and you should treat yourself as a beginner for that portion and return to beginner practice strategies (repetitive, massed practice). This corresponds to Pete Wernick's advice of 'looping' - or repetitively practicing a passage, lick or phrase without stopping - to get the hang of what your fingers need to do. Then you can return to your random practice schedule of the entire skill containing the now former rough spot.

There is a large research literature on 'part' versus 'whole' practice that addresses these issues of breaking down skills for practice into smaller parts. Much of this is common sense in practicing music. The motor commands for performing a skill that is comprised of several movements produced in rapid, continuous succession (like a lick or phrase) are different than the motor commands you would use if you produced each movement with an intervening pause. In other words, the motor commands needed to produce a lick aren't simply a sped-up version of the individual strokes. While you might want to move slowly through a strange new lick at first to learn where your fingers go, you aren't really practicing the needed motor commands until your movements are occurring fast enough that the notes (and movements) are flowing smoothly from one-to-another (which occurs at a relatively slow speed, so don't mistake this as advice to speed up your practice prematurely).

The rule to apply for 'part' practice is to quickly group the elements together into larger units so that you can practice the motor commands for phrases (e.g., 2, 4 or 8 beat groupings), rather than individual notes. Likewise, the transition from one phrase to another may be tricky because you are changing left-hand position, or because the second phrase may require that you lead with a particular finger while habit wants to call on a different finger against your bidding. In this case, the part you need to practice is the transition from one phrase into the next phrase, not the two phrases separately.

I am confident that these practices, culled from an extensive research literature, will increase practice efficiency, particularly for those of us on a limited practice budget. For those of you who have an unlimited practice time each day, you also will advance quicker by following these principles.

Keep on Pickin'
Angie's Banjo.com

1 Comments:

At 2:50 AM, banjo brian said...

WOW! what a great article! I am a relatively new banjo picker and have been trying to teach myself using video's and books, this will be a great help.

 

Post a Comment

Links to this post:

Create a Link

<< Home