Saturday, February 28, 2009

V . . ual . zat . on

Phaedrus said:

I will leave the question open for now if I think that Visualization is the key for further improvement for me.

But if it is, I do not know if there is an easy fix. Yet if there is, and you would help me find it, I will carry you on my shoulders through the streets of your town to a bar where closing hour is only declared when the last customer stumbles home, or when the beer has run out!

LOL. I hope he has a strong back:)

Let's see if we can earn some beer. First we have to make an educated guess how big the chance is that visualisation alone can be the core of the problem. So far I have identified 3 tasks that have to be performed simultaneously:

  • Pattern recognition.
  • Evaluation.
  • Visualization.
From these, two tasks are performed best in an unconscious automatic matter, so that they don't put strain on the short term memory (STM). These tasks are pattern recognition and visualization. Evaluation, thinking and guiding the line of thought is done best consciously.

Training should bring about the transfer of conscious performed tasks into the procedural part of the brain. If you perform tasks that are done best unconsciously in a conscious manner, you make heavy use of your STM. That is shown by a tendency to repeat the same actions over and over again, since the information in the STM fades away quickly and it needs constantly to be refreshed.

List of patterns.
For pattern recognition a short narrative is the cue that helps retrieving. See the picture above which is summarized well by the narrative "a car". As you see pattern recognition is so strong that you even can't look at the picture above without seeing a car. So pattern recognition in itself is not the problem. The problem lies in creating a list of items (car, rabbit, ship etc.) to be recognized in clouds. Or in chess positions.
A good way to create such list is to denominate the characteristics of a chess position and to compare your list with the list of a good annotator. Quality, frequency and impact on the game are factors that are paramount for the items on the list.

To enhance your visualization suitable low level drills are needed. To identify what needs drilling I will try to use the following criterium: what I tend to repeat during calculation. Since the tendency to repeat is a clear sign of a fading STM.
Perfect visualization (100%) I call with my eyes open moving pieces on an analysis board.
I estimate my visualization skills at 15% or so. To settle the visualization stuff once and for all I intend to go another mile with it. The pattern recognition department doesn't seem too difficult now I know how to approach it. But visualization means skills.
I cannot give Phaedrus good advice when I haven't tried it myself succesfully so I started again with board visualization exercises. The maximum I can hope for with that is perfect board visualization which equals to playing a chess game on an empty board. Since playing with a visible board (100% visualization) but with invisible pieces is still a very difficult task I can't expect much improvement from it ratingwise, but I hope to get a clear picture of how board vision should be drilled. Since an efficient drilling system would help once I start with the invisible pieces.

Effect of good pattern recognition and bad visualization.
If I translate the words of Phaedrus he claims that he has above average pattern recognition skills (for his level) and under average visualisation skills. What should that mean for his play?
His list of patterns is larger than of his opponent. It probably means his list doesn't contain tactical patterns soleley but positional and endgame patterns too. He is less inclined to find moves by trial and error but instead he recognizes a pattern and "thinks backwards" to find the moves that lead to it. He will not calculate further than the first stepping stone usually. He will avoid complexity and choose quiet solid openings. Is this true?

Another question is, is Phaedrus able to visualization improvement at all? Did he for instance improve while playing Troyis?

1 comment:

  1. Dear Temposchlucker,

    You make a lot of very good points here. I need a few days to give you a proper reaction. I was swamped this week at work. I hope to be able to respond this Sunday, probably with a post on my own blog.

    Thanks for giving me the honor of such a personal and direct response.