I’ve been chatting a lot about this with other folk on the UKU Coaches group, and also reading some textbooks and attending lectures. How do we learn complex motor skills, and how should we practice in order to learn them better?
Complex motor skills, by definition, have lots of things going on at once. Throwing a disc involves hundreds of muscles – basically the whole body – and for a complete beginner there’s no way that you can possibly take conscious control of all of that each time. Therefore, you’re usually far better to undertake what’s called blocked practice – lots and lots of repetitions of the same throw again and again.
What this enables you to do is to simply program your body to do the same as last time, and then adjust one thing at a time. For example, do it again, but keep the elbow lower, or release the disc later, or whatever. Your conscious mind can cope with the few things you change, and your unconscious can just repeat all the other stuff. Slowly, your brain makes sense of the whole movement.
In particular, you need to be repeating the motion immediately to make the most of this kind of learning. Research from other fields suggest that after about 1 second of waiting for something (e.g. someone to throw it back) you start to lose focus, and after 10 seconds your mind is completely elsewhere. Next time you wait for a website to load you’ll see this for yourself… So: use more than one disc when throwing in pairs. Whatever you throw, do it again immediately and make any adjustments you need (more inside out, lower, whatever).
Once you’ve got the basics of throwing, something interesting happens with blocked practice. Research shows that a prolonged session of blocked practice will improve your throws during that session, but that this improvement is not sustained or transferable. You won’t be very good at the start of your next session, and you’ll be terrible at learning new variations like low-release or whatever.
This is for the same reason that blocked practice works in the first place – your brain does not retrieve (from long-term memory) the whole biomechanical ‘program’ for the throw each time, it just tells your body to do what it did last time with small adjustments. So the whole thing doesn’t end up getting into long-term memory very well, and you also don’t learn to separate out the relevant parts (arm movement, body position) in the way that you must if you need to retrieve the whole throw from memory each time.
Just like compressing a computer file to fit in an email, your brain will learn the most efficient way to store the info about the throw if you force it to regularly retrieve it in full. And the most efficient way to store it is to reduce it to ‘chunks’ of info, which then makes the whole movement more readily recalled and more transferable.
To digress a moment – if I gave you a random list of letters and let you look at them for only a few seconds, you would remember about 7 of them, give or take. But if I gave you a list of words, you’d also remember about 7 of them – and that’s a lot more letters! Your short-term memory remembers about seven things, of any type – but if those 7 things are chunks of other things, the make-up of which is already in your long term memory, then you can effectively remember far more. A particular backhand might be stored as ‘arm movement 3’ + ‘leg movement 1’; where arm movement 3 might be made up of ‘wrist movement 4’ + ‘elbow movement 2’ + ‘grip strength 4’, and so on down to very low-level muscle movements – you group the simple but numerous basic muscle movements into higher-order constructs, just as you use letters to make words to make sentences to make paragraphs… and that way you remember them. You can recall the whole thing more quickly, and as an added bonus you’ve generated a whole bunch of independent subroutines for performing parts of the action – which you can transfer to other situations.
So instead of blocked practice, you now need to do random practice, where you regularly switch which throw you’re performing. You have to recall the entire mechanics of the backhand, then forget that and retrieve the forehand program, and so on. That enables you to practice the act of making a complete throw at short notice without making 5 practice throws each time.
In an ideal world, you want to be making things increasingly game-like as you improve: instead of just deciding to throw forehand then backhand, you should react to something else that forces you to quickly choose one throw or the other. Once you’ve got the basics of throwing, you should be doing drills that involve other decisions and force you to vary your throw – hitting a receiver, breaking a force, all that stuff. But even if you’re just throwing around in pairs, you should at least be varying what you do more often, otherwise you won’t be able to reliably recreate the whole throw in a game.
The problem with genuinely random practice is that you learn better how to retrieve your existing throwing motion from memory, but you don’t actually improve it much – you’re not able to focus on the things that aren’t perfect and make them better, because you’re doing a different throw next time.
So again – use 2 discs when throwing around! Whatever you throw, throw it again immediately; but then next time you get the discs, you can throw something different. This gives all the benefits of random practice, in that you have to keep retrieving a different throw from memory; but it also allows you to make small adjustments and improvements if the throw doesn’t come out exactly as you’d like. It also, as it turns out, increases focus – you have to pay attention to the first throw in order to repeat it – and focus is of course the other key thing you need if you’re trying to improve. You don’t get much better if you’re chucking a disc while thinking about what to have for tea…
That’s enough for today… 😉