Learning to finger whistle

I nearly called this post 'What I learned on my holidays,' but reflected that I had learned other things, too, and also that title sounded a little like a junior school essay.

But what I want to reflect on is how I learned to produce that shrill loud, piercing whistle that one produces with one's fingers between one's lips.

I have wanted to be able to do that whistle for years, in a vague kind of way; but never enough to take any actual steps to learn the skill. However, as my dog grows older, harder of hearing and more confused (and you can keep your jokes about dogs growing more like their masters to yourself) it has suddenly become a more pressing need. I love to walk her on the fells, and let her roam freely - but I do need to be able to call her back, and increasingly my old-style whistle just wasn't loud enough.

Mike (my son) can do it, of course, so I asked him to teach me. He told me what to do, (something like this) and demonstrated it; and then I tried - and nothing. 'Now it's just about practice,' he assured me.

So I kept practicing. And eventually, some semblance of a whistle emerged from between my fingers.  

I had thought that once I had found that sweet spot, I would be able to replicate it, and increase volume, pitch and so on by further trial and error. But it has proved to be a more complicated business than that. Four weeks on, and I am still finding that sometimes I put my fingers to my mouth, and no noise emerges. Further, I don't know what to do when that happens. I don't know what makes the noise that does (unreliably) emerge higher or lower pitched. In fact, I don't really have any understanding of the skill I am practicing.

But nonetheless, practice works. The frequency of getting no noise at all is reducing; and the frequency of getting a really good noise is increasing (I'm up to about 90% of the time, now).

Which raises the fascinating question, for those interested in learning: how does that work? Why does practice work, when I don't know what is making the difference?  Clearly, I am gaining feedback each time I practice: either sound emerges or it doesn't.  But my inability to recognise what is making the difference means that I can't (consciously) try to to it better next time. Yet, nonetheless, over time, the more I practice, the better I get.

It seems to me that this experience raises serious questions for our models of learning: I am approaching unconscious competence without going through conscious competence: is that even allowed? And does it make (for example) sports coaching redundant?  

 Maybe the best way to learn a forehand smash at tennis is just to try lots and lots of them, without a coach telling you what to do differently each time. Perhaps your body will learn the skill more quickly like that?  I'd like to think not: I prefer to believe that some guidance and instruction really do add value. But I'd be even more interested to know what evidence there is to believe that, and what theory can be built around these different routes to acquiring skill. 'Muscle memory' is a label for this - a potentially misleading one, as no memory resides in the muscles; but even when not misleading, it seems to me to describe rather than explain...