Sunday, February 17, 2013

Practice makes... less errors.

Oof, its been a big gap since my last post. And I can only blame one thing, that's been a severe bane on all of my productivity: I re-activated my World of Warcraft account. See, they were having a promotion, and I got free new level 80 character, and the expansion for free, a week of free game time, and yeah... a moment of weakness. Worst idea ever. But I just can't stop.

However, I hope to make it up with a particularly interesting article I found a few months ago before I self-sabotaged my social life.

Error Reduction During Practice: A Novel Method for Learning to Kick Free-Kicks in Soccer, by Savelsbergh, G., Canal-Bruland, R., and van der Kamp, J. International Journal of Sports Sciene & Coaching, Vol. 7 (1), p. 47-57, 2012.

Error reduction. I don't like making mistakes. Presumably you don't either. The classic adage is that it takes 10 years or 10,000 hours to become truly an expert at a skill (which, I might add, was a concept long before Malcolm Gladwell said it). I'd say conventional wisdom for most tasks is that you become really, really good at something by doing it over and over and over and over and over again, until the mechanics of the skill are so ingrained in your brain and your muscles that you could rattle it off at any place, any where, any time.

Like..... on the wing of a plane.

This study looks at a practice protocol that reduces those errors using progressive task difficulty. When I think of practicing, especially in sports at a high level, it does seem like that we often find a skill difficulty that we want to achieve, and then practice it over and over again. In the sport discussed in this study, soccer, you could think of it like trying to take a sweet free kick that curls over the wall and into the goal. I think most people would envision the best way to practice is to just set up that scenario and do it over and over and over again. That pretty much sums up consistent task difficulty - where you set up a practice at one difficulty, and just do it over and over until you've got it down. The study asks whether that really promotes overall error reduction better than a different method:

"even elite players practice their free-kicks after regular training sessions and devoid of any evidence-based practice protocols. Typically soccer players adopt a practice protocol in which they produce endless reptitions from a constant distance to make the practice as specific as possible relative to the real game scenario. This type of practice, however, is accompanied by many errors."

Do you notice a parallel with orienteering in this statement?

The study used 40 soccer players. They would do a specific shooting test that would test the player's ability at 5 different shot locations, each of increasing difficulty, practice, then do a post-test, and finally a retention-test. The practice was split into four groups, one where they practices with increasing difficultly of shots, one with random difficulty, one with only the most difficult shot, and one where the practice got progressively easier. The results of each shot were used in a point system.

Oooh, decisive.
The results paint a pretty rosy picture of the study's hypothesis. The increasing difficult practice had the most significant improvement in performance, although its also worth noting that the constant difficulty group had improvement, and had an impressive jump in retention as well. The other two groups had no improvement or even negative improvement. So, the authors conclude, there is something to be said for this error reduction practice protocol. But, being scientists, they also hedge their bets, particularly because the traditional constant difficult practice did result in almost identical retention performance. Personally, I would also wonder if the parallels between the practice session and the test session, both being of progressively increasing difficulty style, affected a person's thoughts as well. Did testing the easy shots first make the harder shots easier? Another questions for another day, I digress.

"Nevertheless, the present findings also underscore that it is too premature to replace constant practice protocols."

So, what does this have to do with orienteering?

I think for most people, orienteering is a constant struggle with error reduction. Even for elites, 5 seconds left or 5 seconds right is an error. (Aside: I had a very insightful orienteering point out, also, that its not relevant if the error is 8 minutes long or 10 seconds long, if there is a loss of time, it is an error. Magnitude is irrelevant.) Its almost a case of who can make the least errors, since not winning every single leg could maybe be construed as an error. How do we reduce these errors?

Naturally, we all go out for training courses with the goal of repeatedly hitting each control error-free. Not every control is of the exact same difficultly, so you could probably guess its most similar to the random difficulty practice of the soccer study. Random difficulty actually decreased performance! I don't imagine most coaches would need to hear this, but it certainly seems like it is important to at least maintain as close to a constant difficulty as possible. 

Alternatively, you could also consider ramping up the difficult on training sessions. Say you were doing a corridor-O, you could make the first 1/4 quite easy, and then ramp up the difficulty so that the last 1/4 was the most difficult. No, it doesn't necessarily simulate a race situation, since races don't do that, but based on the above study it may improve all-around error reduction, and particularly at the highest difficulty tasks. 

I don't want to go without mentioning too, that I am aware that many coaches already often do this. I know of one coach that recommended to me once to run the easy easy courses by map memory before going out on my own. More progression!

A "new" idea from 25 years ago!

I'm not sure this is exactly mind-blowing stuff. Plenty of people already have the feeling that they like to "warm-up" for orienteering by doing easier stuff first. Hopefully, though, progressively increasing the difficulty of one's navigation training will now have a new appreciation as a purposeful way to reduce errors. As one noted sport psychologist said, "intention brings attention". You'll have more focus on your task if you have a reason for doing it.

As I was doing research for a side-project, I came across a mini article from the Scientific Journal of Orienteering from 1987. Its by someone from the Universite of Sherbrooke. He theorizes that the optimal way to become more proficient at navigating is to always maintain the running speed you aim to run at, and only introduce progressively more difficult navigation tasks. Only once a person is able to maintain that running speed at one difficulty may the progress to the next one. That's not unlike our study! Progressing the task difficult to become more and more difficult, but in this case maintaining another variable, the running, at an optimal speed. 

Check out the article here: http://orienteering.org/wp-content/uploads/2010/12/Scientific-Journal-of-Orienteering-1987-Vol.31.pdf 

Scroll down to page 93.

Or don't, and be distracted by this, and consider whether they should work on some error-reduction techniques: