[hugh]

Hi,
Is anyone familiar with the work being done by scientists such as Schmidt,Magill and Ericson in the Motor Learning fields?
I am referring to their findings on practice regimes for optimizing learning and performance on demand.
I've read their books but still am a little bumfuddled.
Tia,Hugh

[hugh]

Uhh,
How about some keywords:
Variable practice
Contextual interference
Deliberate practice
Knowledge of performance
Feedback bandwidth
Regulatory conditions
Attentional focus during motor learning
...Ring any bells?
I began researching this stuff a few years ago looking for any science on optimizing training.Turned out there was less than you'd think.But what I did find was interesting enough.If anyone's interested I'll try to summarize what I think I've learned.

[Flexmoney]

Please do.

[raz-0]

Flexmoney, on Jul 16 2007, 07:09 AM, said:

Please do.

seconded.

[mjbine]

I am interested.

[hugh]

Okay,here goes,
First let me state what I was trying to learn:what is the best way to practice a motor skill(shooting,specifically)?As it turns out,when you start looking into it,most motor skill research is directed at two groups:children with developmental problems like Autism or Down's,and adults being rehabilitated after severe CNS injury.
Sports related research is heavily skewed towards fitness goals and team stuff.
Finally I found the work of Dr. Richard Schmidt.He did research on learning and retaining various simple motor skills.Typically He would take three groups and have them practice a simple task.Group one is the control group,group two would practice one way,and group three another.Each group was tested and brought back some time later(a couple of weeks)and retested.
What he discovered was not what he expected.
The way each group trained produced very different results.So different,that when it was checked and retested and other scientists repeated it,it needed a new theory to explain it.It has since been tested on a variety of skills,and from novice to experts.
Suppose you are going to practice three skills today,A,B and C.The commonsense method is to practice all your repetitions of A,then all of B,then all of C.This is called Blocked practice.
On the other hand,suppose that you interleaved the three tasks so that a given task was never repeated twice in a row.For example,the practice order might look thus:A,C,B,C,B,A,B,C,A....this is called Random practice.
What was found was that Blocked practice improved skill execution markedly better during practice than did Random practice.The surprise came when the retention tests were given.Random practice subjects show far better retention of the skill.
The theory is that Random practice forces you to retrieve the skill,IOW,to re-solve the problem.By not not drilling the same task repeatedly,your nervous system "forgets" the solution,and has to generate it anew.For example,if you are learning the multiplication tables,once you have solved 8X4,repeating it is not going to force you to learn.
Blocked practice improves performance during practice.Random practice improves the ability to execute on demand.Of course,in a match we only get one chance to do it right,hence the value of random practice.
There's more ...comments,questions?
Hugh

[BSeevers]

Makes sense to me. I am always scoffing at "My fastest time on xxxxxx" or "I broke the record"
What counts is what you can do on demand, the first time.
Get random.

[enoon]

Most interesting. Would like to read more. Please continue.

[CSEMARTIN]

Hugh,
Welcome to the forums! I am extremely interested in this thread. Please continue, you have my undivided attention.

[rwmagnus]

Hugh Thanks for the post keep it comming if there is more. Makes sense logically to me. I never gave it that much thought but maybe random drill practice is the way to go. Heck at this point any practice is the way to go!

[All-Gator]

Hugh,

If I understand with what we are dealing, it is not a matter of randamization of different tasks in practice but rather varying the levels of a given parameter of a one given task. That is, primarily we are emphasizing the acquisition of a recall schema...not the acquisition of a recognition schema. A concrete example (perhaps) would be that if we were working on target acqisition on the draw, we would vary the speed at which we drew; the starting postion (holstered, hand on gun, gun at high ready, etc.); and/or grip strength. The task remains the same. The randamization would be the sequencing of the various levels of the parameter(s) within the given task.

The recall schema ("muscle memory") would then expand to include better recognition schema (i.e., recognition of consequences) in that we would begin different tasks such as drawing free style, strong hand draw, and off hand draw. Again doing so in a randamized sequence.

Or have I missed the point??

BTW, are you studying kinesiology/psychology/rehab or are you just an interested shooter?

[XRe]

Oh, Lord... too many big words...

A couple of us have been discussing practical application of this stuff in our practice regimens - both dryfire and livefire... Would either of you knowledgeable gents (Hugh or Joe) have any thoughts on that? For instance, with randomization in mind, how would one approach a practice session? Any pointers to similar regimens for other sports (I can adapt)??

[All-Gator]

XRe,

The work to which Hugh was referring falls in the field of Cognitive Learning Theory and it's not inconsistent with Brian Enos' approach/thoughts. With that said, let me defer to Hugh until he responds.

A-G

[XRe]

Consistent with Brian's approach or not, I'm interested in knowing more... Efficient and effective practice means more gains, more quickly, with less effort, so... I'm all about all that

[hugh]

Sorry for the delay,
Gator,I'm just an interested shooter.
In your post you are describing what I understand to be Variable Practice.Whereas in Random Practice one changes skills on each rep,in Variable Practice one changes the same skill's parameters.For example,on each rep one could vary the range to the target and/or the number of shots fired.Practicing the same drill repeatedly is known as Constant Practice
Variable Practice makes the execution of the skill more robust because it is practiced in a wide variety of contexts.To extend the math analogy:now we practice solving 8x7,8x2,8x11 etc.
Random practice improves the ON DEMAND execution of the skill by spacing reps of a skill enough for "forgetting" to occur.Back to math:we will practice 8x4 then 3x6,9x5,7x7,2x1...by the time we practice 8x4 again,we have to summon up the solution,not just parrot it.Both Random and Variable have their place.

Contextual Interference (C.I)
Blocked/Random and Constant/Variable practice schemes fall along a continuum called Contextual Interference.C.I. describes the interference with learning caused by changing skills or parameters.As noted earlier,this interference actually improves learning.
C.I. can be modulated: beginning at the low end,there is Blocked/Constant.I think it would be safe to say that this is how most of us practice.We practice a Block of the same drills using the same parameters until we are satisfied.Then a new drill or a new skill is practiced.
1)Draw and fire 2 shots at 7 yards freestyle for 10 reps
2)Draw and fire 3 shots at 10 yards strong hand only for 10 reps
3)At high ready,switch to weak hand only and fire 1 shot at 15 yards for 10 reps

At the highest level of C.I. there is Random/Variable.Here ,on every rep you would change the skill set being practiced-and you would vary the parameters each time a skill set was practiced.
1)Draw and fire 2 shots at 7 yards freestyle
2)Draw and fire 3 shots at 10 yards strong hand only
3)At high ready,switch to weak hand only and fire 1 shot at 15 yards
4)Draw and fire 3 shots at 12 yards freestyle
5)Draw and fire 1 shot strong hand only at 5 yards
6)At high ready,switch to weak hand only and fire 2 shots at 8 yards
Etcetera.
XRe,
The best way I have found to design practices using C.I. is to write down the skills that you need to practice.Then, for each skill,list the parameters you will use it in,i.e., freestyle prone at 50?Yeah.Weak hand only at 50?Nah.

Skill Examples:
Shooting on the move
Freestyle
Strong hand only
Weak hand only
Reloads

Parameter Examples:
Distance to target,target difficulty,no shoots,movement
Number of Shots
Par times
Starting positions(Draw,gun in hand, gun on table)

Prioritize skills and parameters(Saul Kirsch has an excellent template in Thinking Practical Shooting) and then make a list of interleaved drills.Base the list on available time/ammo and there you have it.Modulate the amount of C.I. to avoid burnout.
Sometimes you gotta plink .

Deliberate Practice
K.Anders Ericsson coined this term in his studies on expertise.He studied musicians and chess players at various skill levels and found marked qualitative and quantitative differences in the way that experts practice.Briefly,given a similar amount of practice time,experts spent more time actually practicing,practiced outside their comfort zone,and attacked weaknesses.

This is not an easy way to practice.It requires planning a session thoroughly,and you are committing to that plan.It is best done alone,because a bit of concentration is involved.You must accept lower practice performance or face enormous frustration(remember,you are only getting 1 chance at a drill:ichi-Go,ichi-E )
It is not fun and relaxing.At all.At the end of one session that involved movement in all directions;swingers and multiples;and shooting freestyle, strong-and weak hand only;my nervous system felt ...stretched out.The quality of mindfulness and attention required is tiring.

References
These are the most useful I've found
Motor Learning and Performance by Richard Schmidt
Motor Learning by Richard Magill
Skill Acquisition in Sport edited by Williams and Hodges
Motor Learning and Control for Practitioners by Cheryl Coker(probably most concise and accessible)
The Road To Excellence edited by K.Anders Ericsson
Expert Performance in Sports edited by Starkes and Ericsson
Attention and Motor Skill Learning by Gabriele Wulf

Hugh

[CSEMARTIN]

hugh, on Jul 29 2007, 05:19 PM, said:

experts spent more time actually practicing,practiced outside their comfort zone,and attacked weaknesses.
Hugh

I think there is a lot of truth in that statement. Thanks for posting Hugh.

[enoon]

Thank you again. C.I. explanation and examples answered questions from your first post.

[nphd2000]

Brings me back to the thought of flash cards in school. I can't believe I can remember back that far. I wonder if teachers still use them?

[XRe]

Hugh - thanks for the thoughts, and especially the references.

I'm familiar with Saul's training prioritization template - I use it all the time, actually

I still have some questions in my head - realizing it will take me a while to get through some reading material, perhaps you have some thoughts that might help get started in the right direction? Before I go there, though, understand that I know a complete training plan must have flexibility in it, to account for unforseen things, etc. So, nothing is ever entirely concrete - and sometimes the "plan" for that practice needs to be superceded by other elements, if that becomes apparent during the practice session...

Ok, so that stated - it would appear that there's a time for Blocked/Constant practice, correct? For instance, when first learning the movements of a new skill, or emphasizing a change on an old skill (especially when doing extremely slow "tai chi speed" versions), it would appear that repetition would be useful for a period of time, to insure that the motion is somewhat learned. Following a small period of that, movement down the continuum to a more Random/Variable method will cement and harden the skill for use in performance. Am I following that correctly? A primary practice methodology leaning more towards Random/Variable would be desirable.

What about the circumstance where we find ourselves working in an R/V fashion, attacking the weaknesses (which I'm totally in support of), and we determine that a skill is weak to the point that we cannot execute it? Normally, I'd stop, and work in a Blocked/Constant fashion to refine the skill and find a point where I can now execute it at at least some level. Then return to the practice session as before. To use the math example, what if I discover that I totally have forgotten my 5 times table - I can do the others, 4x, 6x, etc, but 5s are just killing me.... stop and focus on the 5x table for a bit, or???

[hugh]

Dave,
As I understand it,you are right on.Reversion to Blocked /Constant would be the remedy for an unsat skill level or for new learning.If you can determine that a skill has a particular weakness or choke point,you could also employ Part/Whole strategy.This is de-constructing the skill into "chunks"and practicing them separately.As each chunk is smoothed out,re-integrate them into a Whole again.
Since so many shooting skills are ballistic and executed in multiple parallel tasks,Part/Whole practice is used a lot,I think,by shooters intuiting the strategy.
Think of the C.I. continuum as a multiple solution strategy for practice,where you modulate the amount of C.I. for your specific needs.You shift to the learning strategy that fits.You are Stair-stepping in practice up a spiral staircase.
LOHF
Another use for B/C was talked about by Brian with his Limit Of Human Function tests.He would repeat a given exercise until he could do it blindfolded.He would then record his best performance and use it as a benchmark.Periodically he would run the same drill "cold"and record the difference.
His observation was that the LOHF scores didn't vary so much as the "cold"scores improved.

HTH,
Hugh

[XRe]

Thanks, Hugh...

hugh, on Jul 30 2007, 03:41 PM, said:

...you could also employ Part/Whole strategy.This is de-constructing the skill into "chunks"and practicing them separately.As each chunk is smoothed out,re-integrate them into a Whole again.
Since so many shooting skills are ballistic and executed in multiple parallel tasks,Part/Whole practice is used a lot,I think,by shooters intuiting the strategy.

I agree - not knowing that there was an "official" terminology for it, I've been using Part/Whole for a long time, especially for the more complex skills in the game - draws and reloads (especially reloads). It makes a huge difference to be able to break, say, a reload down into fundamental parts and focus on the part of it that's really giving you the most trouble, and then put the pieces back together from there. I've seen huge improvements over the years - and teaching someone to draw w/ Part/Whole makes the whole thing much more fluid for them....

Quote

His observation was that the LOHF scores didn't vary so much as the "cold"scores improved.

And the cold scores are, of course, what really counts... And that's what really caught my attention about this thread....

Cool... time to go get a book or three and delve into it...

Thanks!

Dave

[konkapot]

Probably some of the best information I've ever read here.

FY42385

[hugh]

Thanks,Konkapot!

Hugh

[jostein jensen]

this thing should be pinned or in faq or something...

[murkish]

jostein jensen, on Aug 19 2007, 11:58 AM, said:

this thing should be pinned or in faq or something...

+100