Buy one at Walmart.


I just volunteered to go from just having learned a form last week, to being at demonstration level in two weeks.

You have to have a form down at %60 proficiency to test on it, I hear.

%60 is not demonstration level.  So, the form that I next test on has to be so good that anyone watching just that form has to be convinced that I have tested on it and passed.

In fact, it has to be so polished that anyone watching just that form has to be convinced that I’ve probably got the next form at least up to %60.

In short, I have to do this one form I learned a week ago so well that a master from another school, while watching me, can imagine on the basis of that form that I’m ready for my dan test.

No pressure.

Master Gimp recommends lots of Diet Coke.
Master Gimp recommends lots of Diet Coke.

So, the one thing we know about the Logic Monkey is that the Logic Monkey is lazy, and the Logic Monkey cheats.  How does one cheat at rocking a form?  This is a good question, and one I’m sad I haven’t asked earlier, because if I pull this off, I will have learned more than just form 7.  I will have learned how to crank up my kung fu in general.

Here’s my working theory:

Common knowledge is you have to do the thing one or two hundred times to get it down, and five or six hundred times to rock it.  But I dug up some research on motor memory (what some call muscle memory, but that term has multiple meanings) to see if I could find some optimal timing and schedule.

Here’s what I learned.  First, I learned that short sessions spaced out over several days are more effective than long sessions.  Second, I learned that once you get a move right, you can reinforce those neurons by practicing the correct motion, but there’s swiftly a point of diminishing returns.  Your brain needs a break — sleep, specifically — to go in and reconstruct the wiring between neurons in order to make it automatic.

The ideal, then, is to practice until you get it right, then go a few more times, then get some sleep.  Then test yourself.

Testing yourself is important.  We learn better through testing than review.  Master K.’s “Regard every training session as a test,” is not just a way to motivate us to do our best.  If you are actively trying to remember a thing, rather than simply trying to repeat a thing, your brain gets the message that it really needs to focus on this.

The result?  To get more done in less time, the ideal process seems to be this:

  • Learn the thing.
  • Test yourself on the thing immediately.
  • Did you get it right?  No?  Review.  Test again.
  • Once you’ve gotten it right, maybe, maybe do it one or two more times.
  • Give yourself a reward.  Does the name Pavlov ring a bell?
  • Immediately switch to something different.  The more different it is, the better.  You want to focus all of your attention on something that is very much not the thing you are trying to learn.
  • After you finish your hiatus, test yourself on the thing again.  Do not go review it first.  Test first, then review to see if you got it right.  If you did, new cookie!
  • Repeat the process of switching up until you do it right the first time you test.
  • Leave it alone for the day.
  • Test yourself again in the morning.

Overall, this will be fewer repetitions than the standard method, but they will be harder to do.

So, this is based on a mixture of well-tested knowledge of the learning process, and little-known experiments specifically on motor-learning that aren’t so well tested.

So I’m going to hedge my bets and do this in a way that will get me more reps even though I am trying to use a low-rep, high stress learning model.

I’ll let you in on more of my secret plan tomorrow.  And at the end of next week, we shall see how well my theory holds.  Muahahah.