Monday, April 02, 2012

The essence of transfer

Transfer of the solution from one problem to several other similar problems is based on the mental ability to recognize lookalikes. It cannot be otherwise. All the methods of transfer I mentioned in the previous post share this mental ability as underlying principle. The mental ability to recognize lookalikes = pattern recognition.

So all transfer must in essence be based on pattern recognition. The only thing that differs between the several methods of transfer is the type of pattern. A pattern can be geometrical or textual or conceptual for instance. Only the ability to find lookalikes can multiply the solution from one problem to another.

I have analyzed the questions I ask myself during the phase of "adding intelligence". I can divide them in two: questions that relate to the geometrical patterns of the squares and questions that relate to the thoughtprocess during solution.

As you might remember I distinguished between two moments when to add intelligence:
• During storage of a pattern (= automating intelligence)
• During the solution of a problem (=conscious thinking)
For transfer I only consider the first moment to be relevant.

Summarizing:
• All transfer is based on pattern recognition. Only the type of pattern might differ.
• All transfer is realized at the moment of storage of a pattern.
• CT favours the geometrical patterns of the pieces as type of pattern (plan A)
• My analysis of plan B boils down in the end to the geometrical patterns of the squares.
Which leads to the following question: how can we learn the invisible patterns of the squares?
I realize I have made a few quantum leaps, so I hope you can follow.

To get an idea:

.
.
.
.
White to move.
It is 1900 rated problem. You find the solution here.
You can find the best solution when you are able to identify the focal points of the position:
f8, a3, f3, d1.
During storage time we should find a method to make this invisible pattern explicit.

1. The next step would be: to find a few ( starting with 1 of cause ) "related" problems wich could be solved by knowing/understanding this one.
These rela0ted problems cant be tooo difficult to find, if the number of "pattern" is realy ~400.
Next step then is: to find the path from one solution to the other solution(s)

Maybe easier: analyse problem for problem and find a classification-system on the fly.

It might be easier to see the differences first and get an idea of the classes later.

You did start your analysis with the focal points. maybe you can continue this start by just counting the focal points. Starting with problems with 1 focalpoint then analysing problems with 2 focalpoints asf. Its like a puzzle where you fist sort for color and borderpieces..

2. I could not find the solution and picked it up after maybe 3-5 minutes.

I was aware that a bishop on a3 would actually give checkmate (after I advanced the pawn to f7 of course).

But I also saw, that the black queen is covering the a3 square. I didnt find the side way move of the queen. I know of an easy puzzle though where I do a side way move to a3, give check and win a rook on b2.
I guess the lookalike is not enough.
Anyway, there is a problem in activating my knowledge.
It is like I know thousands of foreign language words passively, but I use only a few hundred in my actively use language.
I know how to make my vocabulary more active, but I dont know how to do this with chess patterns. Maybe a chance could be: after I failed a difficult puzzles like this one, then I try to find a similar pattern I know from my easy puzzles. Somehow trying to connect the two in my brain.
An other method could be in repeating the easy puzzles again and again until the easy puzzles I pretty clear in my mind.
Probably I need to do both. But anyhow, it is difficult to activate known puzzles that only have some similarity with the puzzle I try to solve.

3. @Aox,
good points! I have to think about that.

4. @Munich,
of course I haven't a method yet. I'm already glad I was able to define a new area of importance (focal points).

You are able to find the focal point f8. From there you were able to find the focal point a3. a3 derives its importance from the connection with f8 and the attacker Bc1. Focal f3 derives its importance from the connection with focal points a3 and f7/f8 and the attacker Qf1.

At storage time we should identify these focal points and store them.

After storing focal points for some time I expect we are going to see them easier in new problems.

5. Here I have a good example of a know pattern we all know. And maybe tempo and Aox will solve it pretty fast:
http://chesstempo.com/chess-tactics.html

But it took me 52 seconds to find it.
Why? I am sure I have seen this kind of tactic several times before. But the transfer from my memory to concious thinking was slow.

After I found the solution, I knew I had seen similar puzzles before. But why do I often recall my known puzzles AFTERWARDS?
52 seconds was too slow and I received a yellow flag, hence costing me rating points. Nevertheless, I should have aced it very well. I should have solved it within 10 seconds. But I didnt.

The essence of transfer: I have no good idea how to improve this.
My hope is: If I know at least afterwards it was a very well known pattern, then I start connecting the known puzzle with this new puzzle, and next time I will be able to recall it faster. In case I fail a puzzle or do it too slow, I should think hard about if I had seen a similar puzzle before. I probably had, but I need to search hard in my memory. Often enough I dont recall such a puzzle after I failed puzzle. But probably I only dont recall the pattern, because I am not long enough thinking about it. If I would think about it for 1-2 minutes, chances are that i find a similar pattern burried deep down in my LTM.

What do you think of that?

6. @Munich,

7. Sorry. That one:
http://chesstempo.com/chess-problems/12469

8. @Munich,
it took me an awfull lot of time too. I saw Qf8 as the refutation of Qb8+ and didn't look any further. You see what you expect to see. I don't know how to work around this other than to guide your attention with a thoughtprocess.

9. comforts me to hear it took you long, too.
I am very sure I solve at least 10 (probably more) easy puzzles in my distraction/overloading set 1300-1475, where you play Bxh2/Bxh7 and win the Queen or Rook on Rf8/Rf1/Qf8/Qf1 by taking it with a rook or queen.

They are really the same difficulty, but there rating is between 1300-1475 while this one is not. I cant see anything that rectifies the rating of this one (about 1700). It is really the same tactic/pattern. And I am once more disappointed that I have this pattern in my memory, but the recall gets triggered AFTERWARDS, after I found the solution. Of course it doesnt really happen afterwards. I believe it just took my memory so long to deliver it into my concious thinking. How can I make this transfer faster? I did what could be done: I did a tag sorted set, that showed me many examples of this pattern. All of them I can probably do within 10 seconds. But a slight unimportant change in the set up - and it took me 52 seconds.

As you said in this chapter, it is about the "essence of transfer". And I dont think a plan B would help me here to find it within 10 seconds. Maybe I derived it even with a plan B, and the Plan A just didnt get triggered.

I have no solution for this problem. I simply hope that beeing more and more aware of this pattern, to recall it again and again and thinking about it like I do now - all this will help me not to miss this pattern again (or to need almost a minute for it).