The Steel Guitar Forum

I think this is a good place to jump off into a psuedo-scientific observation I made a few years back about the process of learning. I think it's relevant because the pedal steel takes TONS of learning. Bear with me, this really is interesting. (Yeah, right, Bill.)

About 6 years ago, I attended an informal seminar about neural networks. If you're not familiar with neurons, your brain is made up of zillions of them. Thought and memory are really just synapses (bio switches) making and breaking connections between different neurons in your head. Someone who is a great steel player is someone who got their neurons wired right for the job. But how?

At the seminar, they spoke of an artificial neural network they had devised in software in a computer. It had just one job: to back a little box-like image of a semi truck up to a loading dock. It had no idea how, it just was put to the task and allowed to run. It made rediculous mistakes as it tried and tried. Slowly, it got better. Eventually, it could back the tractor and single trailer rig up to the loading dock from ANY starting position in a single motion, even from a jack-knifed start. Then they added a second trailer and let the neural network learn. Eventually, it got it. That's a maneuver which is impossible for humans. Then they added a third trailer. It got that eventually, too. It was amazing.

There was one key element in this learning process: <u>mistakes</u>. The neural network had to try and blow it and blow it again and again countless times. It made rediculous attempts at first and failed miserably, but eventually, after countless errors and slow but stead progress, it achieved perfection.

This was a moment of enlightment for me. We say that "we all make mistakes" as we learn and live life, but by and large mistakes are viewed as generally undesirable or even bad. I grew up just hating to make mistakes. But this artificial neural network lesson taught me that the brain, a natural neural network, <u>needs</u> to make mistakes. Mistakes and failure aren't bad things, they're <u>good</u>. Egads. "We all learn from our mistakes" is a phrase we've all heard a million times. But this seminar made the mechanism of that truth startlingly clear to me.

How does this apply to pedal steel? Well.... I'm still a steel beginner, and I'm going to go on trying to play pedal steel. Which means I'll go on making tons of mistakes. Which means I'm learning.

(There! Now the Forum has covered everything from cabinet drop to brain science!)

------------------
<font size=-1>Bill (steel player impersonator) | MSA Classic U12 | Email | My music | Steeler birthdays | Over 50?</font>

So Bill, You have finnaly figured it out! Now where I get one of those "nuerons" I need all the help I can get!
I haven,t even got my "burnt up MSA" fixed up yet. See! Ole Dave Frye

Bill,

I have to say, that was one of the most thought provoking posts I have seen to date! Because I started playing in the early 60’s, when there was very little teaching “material” out there, no “Forum” on the internet, and I couldn’t afford (or didn’t have enough sense!) to go to one of the 2 people in San Antonio that actually taught Steel guitar…I became an “ear” player. Don’t know any music, other than what little I have learned over the years, and I learned it all on a bandstand. I guess that is why your thoughts hit home.

Mistakes?! Good grief, if I had a nickel for everyone I have made I would be right up there with Bill Gates in wealth! And trust me, my monetary wealth would still be growing every weekend. I would take exception with one thing you said however…

“We all learn by our mistakes”. Sadly enough, for whatever reasons, there are people who do not learn from their mistakes. I have no idea why. Maybe they don’t have enough “mistake” neurons in their brain to allow them to admit they actually made one. However, you are correct. We all do make them, it is no big crime to do so, and those of us who will succeed WILL learn by those mistakes.

I am reminded of a brief conversation I had with the first Sales Manager I ever worked for in the radio business years ago. I sat around the office for about a week or ten days trying to figure out what to do, riding around on calls with the experienced sales people, asking questions, etc. One day the Sales Manager walked in and said, “I want you out on the street tomorrow morning at 9 AM making calls, and I don’t want to see your smiling face in here until at least 4 in the afternoon!” I said, “Oh no…I don’t know enough and I would make a lot of mistakes!” He leaned across my desk and got right in my face and said, “Of course you will stupid! And the sooner you make those mistakes, and learn from them, the sooner you will start making ME money. By the way…if that doesn’t happen fairly soon, you won’t have this job!”

Guess who went out and started making calls the next day?

B. Bailey Brown

<SMALL>But this artificial neural network lesson taught me that the brain, a natural neural network, needs to make mistakes. Mistakes and failure aren't bad things, they're good.</SMALL>

...Bill, I'm just glad you aren't a surgeon working on my natural neural network...

THAT was a great post! Puts it all in perspective.

Rick

Neurosurgeon? Never. I have a hard enough time cutting up notes a steel guitar.<FONT SIZE=1 COLOR="#8e236b"><p align=CENTER>[This message was edited by Bill Llewellyn on 28 September 2001 at 07:35 AM.]</p></FONT>

Very interesting. I've often said if I could just remember all the things I forgot instead of all the things I know I'd be twice the musician I am now. How's that for food for thought?? Ziggie

Give me a break!

There is a school of thought that goes as follows: When learning to play a piece of music, on any instrument, the most efficient way is to play it correctly the first time, and every time. If we play it slowly enough and carefully enough we can do this.(If I never play it wrong,I will always play it right.) Every time we make a mistake during the learning process we do indeed learn that mistake to some degree, and it is liable to reappear at the wrong moment. Put another way, the process of learning to play a musical instrument means acquiring habits, and a habit is acquired by repeating an action in exactly the same way for as long as it takes. Any deviation is a retrograde step. Failure to realize this is the reason why some people make slow progress on an instrument, they in fact practise mistakes. The foregoing is a point of view held by some teachers, what do you think?

Bruce: that seams...uhhh seems to make real good cents...umm...scents...errr....sense!

Bruce:
I agree with what you state... I just don't always follow the regimen...

Joe Wright expressed something like this it in one of his seminars -

"Practice doesn't make Perfect- Perfect Practice Makes Perfect"

<SMALL>Any deviation is a retrograde step. Failure to realize this is the reason why some people make slow progress on an instrument, they in fact practise mistakes. The foregoing is a point of view held by some teachers, what do you think?</SMALL>

Bruce, your idea is pretty much what I remember from Howard Roberts, who was considered an expert on the process of learning guitar. Start with a new idea as slow as you have to in order to play it correctly, and if you make a mistake, stop immediately and correct it.

Not to put too fine a point on it, but I get a little uncomfortable with the implication, assumed but not yet proven by many of today's brain scientists, that all we are is a biological computer. I think the eventual results of this materialistic view are troubling. I doubt Bill believes in this view of man totally.

For many, the creative process may seem ordinary and explainable this way, but that hasn't been my experience. This is why I thought that, although Howard Roberts had a lot of very useful theories, they were based on a limited or partial view of what we are.

Jeff <FONT SIZE=1 COLOR="#8e236b"><p align=CENTER>[This message was edited by Jeff A. Smith on 28 September 2001 at 02:50 PM.]</p></FONT>

Jeepers, Bruce, that's an interesting point.

Mmmm.....

I think it's fair to assume that doing something right the first time is really the proper assembling of (perhaps unrelated) habits or patterns learned previously into a string of actions which properly accomplish a new result. Where did the previously learned habits and patterns develop? From perfect first time executions? I'd guess they came to be via trial and error. Just watch a baby try to talk or walk. Hundreds or even thousands of attempts before they get it. But once they learn a building block vocabulary, they can say at least some brand new words perfectly the first time. Once they learn to take steps or move about in varied fashion, they can do simple dance steps right the first time. I would therefore suggest that in order to see how folks do something right the first time, you still have to go back to the way the foundation was laid, and that harkens back to skills acquired via trial and error.

Try another example. Take a very accomplished driver of a regular motor car. Put him on the track at Indy in a race car and he goes around the loop at high speed flawlessly the first time. Should we put him right into a race? Probably not. He needs to learn what to do when things go wrong. Learning by trial and error is adapive learning. One needs to know what to do when things go wrong. On the steel, when you know you're headed for a train wreck, doesn't it pay to have been there before to know how to steer out of it?

I'd never advocate that doing something the wrong way as a form of perfecting onself is a good idea. But the brain is an adaptive machine. In order to adapt, it needs experience. And mistakes can (and I believe do) provide it with experiences useful for future error prevention.

I don't know if I'd feel comfortable with a teacher who wanted me to accomplish a task right the very first time, unless it were actually a re-assembling of previously perfected habits into a new string of actions. I think that may be what the school of thought Bruce mentions is actually doing.

I guess the root of my original assertion is that people need the freedom to blow it from time to time. Based on the way the brain seems to be built, it's part of learning. But we have a natural human tendancy to frown on, be ashamed of, or be impatient with the error part of trial-and-error, which can hamper learning. Maybe my message was mostly for perfectionists like myself who, if we don't get it right on the first try, want to take our marbles and go home.

Interesting, Bill.

But a computer can attempt a task millions and millions of times non-stop until it gets it right, and humans can't do that. The computer could probably try backing that truck up as many times in a day as I could attempt the end of "Raisin' the Dickens" in my whole lifetime.

Of course, backng up three trailers is easier than "Raisin' the Dickens"...

C#

Good point, Carl. The computer model driving the truck had just one neuron, though. The human brain, on the other hand, has an enormous number. The human brain can also draw from a wealth of past experience (I suppose intuition would be one form of such) for much faster adapataion than an experience-less computer model. Again, that's the entry point for previously perfected skills building into a new skill.

Please don't read me wrong. I honestly don't view the human mind as a mere computer at all, though it does compute. I see it as an enormously complicated marvel made by our Creator, so I take its expression of learning, creativity, cognizanse, and so forth as being far greater than the sum of its parts. Isn't it great?

I guess the discussion boils down to this: Are mistakes necessary for adequate learning? I would say that (apart from instinct, and that's a whole 'nother discussion) if one borrows from previously developed skill sets, then quite possibly not. But I also tend to think the building blocks of behavior and skill (pedal steel included) rest upon a process of having tested different paths, learning and choosing the right ones based on assorted successes and some failures, and committing the successful ones to permenance.

With me and the pedal steel, it's pretty evident I don't have a lot of the aforementioned sub-skills needed to be very good at this point, or to learn very fast. Manual dexterity is not one of my strengths, and I think that's really important on steel. So it's an uphill battle. However, I allow for trial and error to be part of the learning process, assuming error is normal--as long as I get the feedback that an error is indeed an error. If I don't, I rehearse the error and I get into the rut alluded to by Joe Wright. In fact, I think that's a really important issue not previously mentioned.... In leaning by trial and error, proper corrective feedback is crucial.

Isn't this fun? <FONT SIZE=1 COLOR="#8e236b"><p align=CENTER>[This message was edited by Bill Llewellyn on 28 September 2001 at 04:34 PM.]</p></FONT>

I think the hardest part of learning to play Pedal Steel is "learning when we ARE making mistakes". NEVER get "too cocky" thinking "I've got it now"! If you think you are "there" then you've just stopped yourself on the ladder up to being an even better player. We all make mistakes, heck, I even left mistakes on my Cd on purpose. Learn from other peoples mistakes, whatever it takes. Enjoy it all along the way but don't smell the roses too long or you'll get allergies. IMHO
Dave

Yeah, good post. Try learning PSG while you are in great pain, taking pain killers (the HARD stuff) like candy in between surgeries on your rotator cuffs and see how much you remember when you get home and try to practice for the week. I had to resort to taking notes so I could try and remember what was said. I wonder where I'd be now (after 10 months) if I had started with a whole brain instead of a doped up half a brain. Well, it can only get better!! (WE HOPE) Kinda funny when your teacher looks at you and says things like-- should you be DRIVING??

Those were ALL good post..Steel Philosophy 101. As a former carreer Air Force guy..I know about "Do It Right The First Time", "Zero Defects", etc etc.. What about talent or natural talent?. One must Have a natural talent for something...A person has a nitch...All people do not have the same "nitch".The nueron theory is valid, however is is only part of the make up of a person.Genetics also play a large part in the development of a person. Trial and error will always play a part in development. It is improbable that any one can go through life without making mistakes! We stive not to..but it is impractical to even think that we can be Perfect..For WE Know that there is only One Perfect Thing..but this is not the place to get into religious philosiphy. If we must learn WITHOUT mistakes, or we learn BY mistakes, why can't all of us obtain the Talent of an Emmons,Charleton,Heughy,Rugg, T. White,L Green,H Walace, J Crawford,P Franklin,S West. or those other top pickers.
I have EVEN SEEN some of those TOP pickers "cringe" or make a small smile .or indication of "Embaresment" when they have MADE A Mistake!!!!Let's face it . We All make mistakes. The best we can do is strive to do our best and eliminate mistakes to the BEST OF OUR ABILITY.That's all that any one can ask for. Talent is still in the equation of life...So let's just go out and pick and have FUN!!!!!Enjoy it while we can..Play to our ability..that's all that we can ask for.
Grouchyvet

Bob..Right on. You say "play to our ability,"
unfortunately many people never manage to do that.They know they need to practise, but they do not know how, and no-one tells them. I write as one who spent too many of his early years practising mistakes. A first reading of Bill's first post and my reply would seem to show opposing ideas, but in fact the two approaches are complimentary, and both are essential to the learning process. First let me say that this post contains nothing new, the essence of what I am about to write can be found in books by eminent teachers and virtuoso performers on various instruments. Faced with a piece of music that presents difficulties, say the first phrase of Three blind mice, or Raisin' the dickens,according to ability, we have choices to make,(neck position, pedals, knees, right hand picking etc.) and this usually means some experimentation, trying out various things until we arrive at what seems to be the best way of playing the phrase.This could be described as making a series of mistakes, but the vital thing is to keep the experimental attitude, not trying to learn the phrase. Once the "how to play it" decision is made, then the "no mistakes" repetition part of the learning process starts. It is important to realize that the part of the brain that does the memorising cannot differentiate, for instance, beween right and wrong notes, it will learn whatever you play! Anyone who read any of the obituaries to Isac Stern last week may have seen the tribute he once paid (in a book ) to his teacher who he said "taught me how to teach myself." Finally, to Bill Fall, thanks for giving me the best laff, er, laugh of the day.

I thought it was Maurice Anderson who said:
"practice does NOT make perfect.
PERFECT practice makes perfect!"I could be wrong.


Wayne

Slick- I believe you are right!.....al

Good discussion...and EXCUSE MY SPELLING..I need Practice!!! or got yo get the right nuerons..opps..there I go again.. We ARE a scientific group!!It's a joy to be a part of this forum.
Grouchyvet

Hey, I really like the concept of perfect practice. Why practice, though, if one is already doing it correctly? Well, you practice to reinforce the right way of doing it. But you did it right the first time. If so, you're already there. If you did not do it right the first time ("right" can mean a host of things, such as fast enough, stucatto enough, and so forth, not just playing the right notes on the very first pass), you played imperfectly, from which you learned. Eureka. That's my point. Therein lies adaptive learning. Imperfections can teach.

BUT.... In order to adapt this way, you have to know you made a mistake and fix it each time. Without corrective feedback (a teacher, good training material, the Forum , etc.), your're potentially lost and adaptive learning breaks down. Then you have imperfect practice which reinforces error and you end up in a rut. Been there, done that! But rutting due to lack of corrective feedback (and the resulting imperfect practice) does not nullify the process of adaptive learning. In fact, it supports the whole point. How so? Because whatever it is you reinforce, you burn in to your head. (Remember the goal of computer based neuron?) You learn, you just learn the wrong thing. (The same thing would have happened if the artificial semi truck in the computer had an indistinct or moving target to back up to.) That makes good teaching sources crucial. You can't have perfect practice (or constructive adaptive learning) if you didn't get shown how to do it right in the first place (as well as on an ongoing basis).

I think a lack of solid, readily available instruction is a real issue for neophyte steelers. Sattelite learners like me are in the unfortunate position of risking much imperfect practice. I adapt to my own mistakes (since I have little of the corrective feedback mentioned above) and burn them in. I remember my first steel lesson. In a few moments, my teacher adjusted my right hand position and opened up anew world for me. I then had to unlearn the bad habits to which I had adapted myself.

I never said Joe Wright was the FIRST to say the bit about "Perfect Practice making Perfect"

Anyone ever experience playing better after a period of laying out from playing?

Recently, I sent my guitar to Sierra for some upgrades to my setup, and it was gone for about 6 weeks. I really misse it.

But when it came back, I noticed that all my practice (which I thought would have been fading) was still there, and in fact, I DO believe, I actually improved, and seemed to leap forward a bit.

This also happened when I was required to travel for a few weeks for my job this summer.

Each time I am away, I return fresh, and even better. The same is true for guitar, with which I have played and experienced the phenomenon for many years. I also believe my years of playing guitar and theory are helping me "cross pollenate" to steel, although it is STILL like learning to walk!

My question is that do the neurons benefit from the rest? or is it my imagination? Do they continue to grow and make connections?

------------------
Bill Fulbright
1998 Sierra U12 7x5; Gibson ES-165; Peavey 50-410
ICQ# 2251620 Bill's Launch Pad