This article was originally posted on my SwordSTUFF blog, and has been re-posted here because it aligns completely with the topic of Games and EA teaching design. (I’ve done more since then, so watch for the follow up.)
This article is going to be a case study of using Constraints Lead Approaches in practice, and what to do if someone does not develop the skills you want out of a game or other training activity. Constraints Lead Approaches (CLA) is a motor learning strategy where there is no explicit instructions provided on what to do, instead constraints are provided and the individual develops a movement solution that best helps meet the constraints. The coach shifts from a role of describing and correcting to observing and designing new constraints to subject the athlete to.
Normally I would probably start with an article about CLA, the science behind it (on SwordSTEM) and how to start implementing it as part of a program. But that’s on hold because doing a properly researched and cited CLA article is a lot of work. Whereas this case study became topical in a discussion today and I can write it off the top of my head. So enjoy.
So what is the Finnish Chicken? There are several different versions that are floating around, but the one I was working with goes as follows.
Both fencers start out of distance in right Vom Tag. As they move closer to each other they are allowed to pick one of two actions:
- Direct cut to the head from the right.
- If they both attack, look at who was moving forward and who was standing still. In just about all cases one person didn’t move their feet and thus were clearly attacking in response to their opponent instead of parrying. Auto-lose.
- Parry against the cut to the head from the right.
- If you can’t tell if someone was parrying or attacking it is also an auto-lose. Make it clearer. This isn’t sparring, it’s a game with rules that you need to follow.
The purpose of this is to make it a very simple game from the blade actions side. (If you need to humor some lefties, do a few rounds using left Vom Tag.)
The real decision revolves around if you should attack or parry. If you attack from too far you will almost certainly be parried, however if you move too close you risk being attacked without sufficient time to parry if they chose to attack you. Thus fencers are constrained with limited actions and should focus on the distance management aspects, figuring out when is the opportune distance and timing to attack or parry.
One would expect that you get a 50-50 ratio of attack and defense wins.
- If the defender is constantly winning then obviously people will stop throwing attacks from so far away, and creep closer. Thus increasing the chance the initial attack will land.
- If the attacker is constantly winning then people will try to throw attacks from further and further away, trying to get their attack in before their opponent does.
Naturally this is going to fluctuate a bit, as all natural systems due. And you can expect some ups and downs as people learn and new strategies propagate. What I did not expect was that the defender seemed to be winning almost every time!
The first step was to actually take a count of who was winning the next time I did Finnish Chicken, to make sure I wasn’t crazy and/or confirmation biasing it. And I was right, there was a significant disparity. The defender was winning twice as often as the attacker!
CLA dictates that rather than provide direct instruction a new scenario with new constraints is implemented to try and find a better movement solution. This requires a lot of careful thought and design work to get right. So what did I do? Naturally I just called them in and gave some direct instruction instead.
Rather than berating them I went with a constructive approach, framing it as a discussion.
“Who do you think is winning more, attacker or defender?”
“So what does that suggest the attacker should be doing?”
One brave person who speaks up: “Get closer?”
“Yes! Let’s try that again, but remember that just about all your attacks are being parried right now, so you’ll have to get uncomfortably close.”
After this rousing pep talk we did a few more rounds of the game. An I’m proud to say that my inspirational direct instruction had… absolutely no effect.
“The body cares very little for the words of the coach” is me paraphrasing a semi-famous quote from a movement science researcher. Who the name of and exact quote I can’t remember at this time (sorry dude). What this means is that while everyone clearly understood what needed to be done at a cognitive level, the actual movements and decisions that happened in the split second they were under pressure were not regulated by this higher level cognitive system. And thus all the instruction in the world wouldn’t make anything better.
Given that Direct Instruction has failed, I of course go back to the drawing board. (At this point I’m also discussing the issue with other coaches as an interesting case study. So assume everything that doesn’t work is them, and everything that works is my idea.) First thing to consider is what if there are simpler fundamental skills of measure and timing that can be developed first.
Breaking down the skills needed, two more simplified games which teach component skills are the Direct Attack and Soviet Foil.
- Direct Attack Drill (https://www.youtube.com/watch?v=ysqm9drma60&t=19s)
- Soviet Foil Drill (https://www.youtube.com/watch?v=AMsm6lWqPJU&t=10s)
Introducing the students to these would be a great step to getting the proper distance management in Finnish Chicken… if they weren’t already games that we play all the time and they are very familiar with. Clearly this is a context issue, rather than a lack of exposure to the underlying concepts.
But what I did do was to attempt some priming the next time Finnish Chicken was on the training program. Rather than jump right into the game, from whatever we were doing before, I would instead build up to it with the more simple versions of distance and timing control. Start with Direct Attack, then do some Soviet Foil, and then some sweet 50-50 Finnish Chicken on my third attempt. Problem solved.
The Fix – Again
Of course you can tell by the fact there is still a bunch of the article left that it didn’t work. Sigh. On to round 4.
CLA is steeped in what is known as the Ecological Approach, a key component of which is that context is everything. So how do I make the context most closely match that which I’m trying to teach?
First of all I changed how I did my Direct Attack. The version I have students to is different from that in the video linked. Rather than parry I use a step back as the defender’s required motion. This is because I’m pretty focused on getting people to move, and avoiding planting their feet. But to help people understand their optimal attacking distance vs the parry, I switched back to the orthodox attack-parry roles.
The second change I made was to do the Direct Attack twice, with different target restrictions each time. The first round was with any upper body target, and the second with only a cut to the head from the right. Naturally the second was much easier to parry, and lead to the two getting far closer before the attack-defense balance evened out.
Finally, some success. It appears working with a more context-specific game was able to get the fencers to act at the appropriate distance, and thus the game equalized out where you would expect it to. (Free vs Ratched explained below)
Other Possible Approaches
This worked well, however there are some other things that could have possibly worked:
Have Them Play More
The first two failed attempts were based on two 10 minute sessions, with only some overlap in participants. This is by far too little time to develop to a stable ‘meta’ of gameplay. It is highly likely that if they continued to play for long enough people would start creeping closer and closer, and it would have the desired effect.
My solution was to keep the same constraints, but do some priming with a different activity beforehand. But the other approach is to change the constraint of the activity itself. One variant that I also considered was the ratchet version. In this neither person can step backwards, only move forward or stand still. My hypothesis was that maybe doing this would force people to be more cognizant of their distance, as steps were now a much more limiting choice and have to be used judiciously. So in my lesson plan for the grand success iteration (also known as Attempt #4 in the chart above) I included the ratchet version after we played the original. The results were fairly similar, but since 50-50 had already been achieved all it did was show that it didn’t produce and skewing over and above the way I had been playing.
And We Lived Happily Ever After
Now that we’ve had a good training session and the correct attacking distance has been internalized there is no need to go through this rigmarole every time we play. So it’s safe to get right into the Finnish Chicken game with no priming before hand….
Which didn’t work at all. Guess I’m trying the ratchet version off the bat next time.