Backward chaining

I recently read the “Strength training and coordination” by Frans Bosch. These are my notes and reflections with regards to kendo from that. All the brilliant stuff is copied from him, all the mistakes and stupid stuff are mine.

Attractors and fluctuators

Complexity of human body makes it necessary to restrict degrees of freedom for efficient movement. Our brain cannot handle the combinatorial explosion of having to figure out a unique pattern of commands for each situation we encounter. Because we clearly can move purposefully something else is going on.

To overcome this combinatorial problem we can divide movement to fluctuators, which are the parts of movement that respond and change according to the specific situation at hand, and attractors which are the more or less same movement patterns or states that stay the same from execution to execution. The gist of the matter is that the brain can command execution of the attractor patterns as a chunk without having to micromanage it at the top level.

When first learning the movement the restrictions the body sets are not necessarily the best with regards to the needs of the context such as ability to adapt to changing conditions, force production or accuracy. Learning in this setting is the process of grooving in optimal attractor states.

The set of optimal attractors depend on the context and also the properties of the central nervous system but also more local signalling and properties of the muscles such as force length curve and attachment points of muscles for leverage. This means that the optimal attractors can fall on more or less the same places between individuals due to our shared general anatomy. On the other hand individual quirks can cause the optimal attractors to be completely different. From a coaching perspective it may be useful to focus on the generally prevalent attractors.

Other point with regards to learning is that the more generally usable the attractor is the more willing the brain is to learn and start to use it. This for instance means that training in variable conditions helps the brain to identify patterns that are widely useful.

Backward chaining

Movement is learned and executed according to a the top down process based on the intention of the movement not by bottom up rules how to move muscles. The external end goal is arrived at by going through stable beacon internal states which allow the brain to chunk the movement in to manageable pieces. At each between state the brain has a, possibly unconscious, goal of getting to the next beacon and eventually to the external end goal. The brain may also monitor the progress of the movement by seeing if we get to the beacon state as expected.

Because attractors are the stable parts of movement it makes sense that the brain uses those as the beacon states. So the theory goes that the brain checks that we are hitting the attractor states and directs us to the end goal via those states using the fluctuating parts to react to deviations from plan.

Because the intention is what the brain cares in the movement when learning a new movement it may be useful to start from the end goal and then work backwards through the intervening beacon states towards the start. This way the intention for the next beacon state and eventually the end goal is clear at each intervening state. Also there typically is more variation in the movement leading up to the end position than in the end position.

Fumikomi movement

In kendo the most common way to attack is using the leap or fumikomi movement. There are other patterns but that seems to be the most common for at least men, kote and tsuki.

What are the stable features i.e. patterns that happen every time in a scoring attack using fumikomi more or less universally among high level players.

Furukawa Kazuo 8. dan scoring men from
  • at shinai contact free leg in air with
  • knee in line or further forward than the ankle and
  • free leg side hip raised and
  • stance leg at around toe off and
  • at launch no forward tilt of hip with regards to stance leg.

The variable features seem to be at least

  • free leg angle prior to raise i.e. whether the front foot slides forward or goes up immediately
  • height of the free leg knee. It seems that there are individual differences here.
  • extension of the stance leg. The stance leg extension can control the distance to target. Typically in kote the stance leg is left more bent than in men.
  • extent of shinai up-down-sideways movement and lever action on hands. Although it is generally accepted that the shinai should make a big arc keeping the center line in practise the shinai typically makes a small vertical movement and may start from any angle.

Backward chaining exercise for fumikomi men

Given the end goal that we want to strike the opponent start from that point and work backwards from there.

  1. train the contact moment: put a bench or a box in front and place the front foot on it. With arms extended in front hit using just fingers and palms while keeping upright. Vary the height of the box, distance from stance leg to the box and the height of the striking target.
  2. as before but vary the amount of up-down-sideways movement of the arms and shinai prior to hitting
  3. launch by starting with both feet on the ground in front of the box while executing the strike ending in the position in step 1. As before variation helps in learning.
Like what you read? Give sakari a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.