A few years make a 9 point difference in this position

[Terry McCracken]

In an experiment with Judith Polgar, they found (with a brainscan) that she actually uses the facial recognition part of the brain to recognize chess positions (/blocks).

Tony

Fascinating. I had suspected that might be the case, but had not heard it was confirmed. I seem to recall GMs can remember something like ten thousand or more chess patterns (AKA chunks). I wonder how many faces a single individual can recall.

It was done pretty funny. They redid some famous experiments. They had Judith sitting on a terras, drinking coffee, when a truck came by with the picture of a chessposition on it (1 by 1 m) . She saw it for about 3 secs.
5 minutes later she was asked if she remembered the position. No Problem.

Then she was told another truck would pass by within 2 minutes. It had a random position. She was unable (though she was warned and had more time) to setup the position afterwards.

Tony

I think that other GM with similiar rating to Judit polgar have better talent for chess and maybe they may do better in memorizing random positions.

I believe that Judit had more support to invest time in chess when she was a child relative to other 2700 GM's.

Uri

Randomised positions are far more difficult to remember as they don't make sense to the brain.

Organized positions do. That is why faces are simple to remember inter alia, is due to symmetry. Asymmetrical stuctures are more confusing to the brain and often the brain will ascribe patterns to these more nebulous shapes distorting what you really see, understand and remember.

An example would be seeing faces or animals etc. on walls or floors with randomised patterns, within clouds this phenomenon is even more pronounced.

Terry

[bob]

I read about this experiment and I simply do not believe in the theory that grandmasters and hobby players get the same in memorizing random positions.

I remember that I read that Bobby Fischer remembered 2 sentences in a language that he never knew.

I believe that the strongest GM's simply have better memory relative to other players and it is not only about chess so it should be also for random positions.

I can also add that chess is also about thinking and not only about memory.

Uri

I guess it's safe to assume that grandmaster at the top have a much better long-term memory than average players.

But anyway, the mentioned study did test two kinds of memory actually. The positions with the positions from real games could be solved with the help of the short-term memory and the help of the long-term memory.
When rebuilding the random positions the players could rely only on their short-term memory as they haven't seen that kind of positions before. But it seems the short-term memory isn't that important in chess play as the grandmasters performed not better than the amateur players when dealing with random positions.

Roman

I wonder if grandmasters were involved in the test and in this case what is the names of the grandmasters.

If you compare masters with weak players then it is possible that you will not find that the masters have better short memory because short memory is not the main factor that can help you at the weak levels but I suspect that strong short memory give additional advantage at the high levels.

Chess is not only about pattern recognition and long memory but also about calculating(at least if we do not talk about blitz).

If you calculate during a game some tree of 100 positions then the ability to remember every position in the tree in order to get conclusions is a clear advantage so it is logical to believe that the strong grandmasters are the people who have this advantage.

Uri

You can find this in "Thought and Choice in Chess" by deGroot

he spent tons of time trying to understand how GM players do what they do, from tracking eye movement, to having them talk as they analyzed, etc...

BTW, this is a book, not a paper...

[bob]

In an experiment with Judith Polgar, they found (with a brainscan) that she actually uses the facial recognition part of the brain to recognize chess positions (/blocks).

Tony

Fascinating. I had suspected that might be the case, but had not heard it was confirmed. I seem to recall GMs can remember something like ten thousand or more chess patterns (AKA chunks). I wonder how many faces a single individual can recall.

It was done pretty funny. They redid some famous experiments. They had Judith sitting on a terras, drinking coffee, when a truck came by with the picture of a chessposition on it (1 by 1 m) . She saw it for about 3 secs.
5 minutes later she was asked if she remembered the position. No Problem.

Then she was told another truck would pass by within 2 minutes. It had a random position. She was unable (though she was warned and had more time) to setup the position afterwards.

Tony

I think that other GM with similiar rating to Judit polgar have better talent for chess and maybe they may do better in memorizing random positions.

I believe that Judit had more support to invest time in chess when she was a child relative to other 2700 GM's.

Uri

Randomised positions are far more difficult to remember as they don't make sense to the brain.

Organized positions do. That is why faces are simple to remember inter alia, is due to symmetry. Asymmetrical stuctures are more confusing to the brain and often the brain will ascribe patterns to these more nebulous shapes distorting what you really see, understand and remember.

An example would be seeing faces or animals etc. on walls or floors with randomised patterns, within clouds this phenomenon is even more pronounced.

Terry

There is a well-quantified "rule" about this... 7 +/- 2

The normal human can remember 7 random things, with an error margin of +/- 2. So some can only get 5, others can get 9, and I am sure there are a few way out on the standard deviation curve that can do somewhat better or worse.

The issue is "what are these things". In a random position, a piece/square is a single thing. In a normal position, the pawn structure might be one thing, the normal positions for pieces, etc. So for a GM the entire position might be a single thing if he has seen it enough.

For example, many of us remember 10 digit phone numbers as 6 pieces of data, the area code, the exchange, and then the final 4 digits are random.

In my case, my phone number is the same as the local auto-zone except the last 4 digits of my number are 4901, while autozone is 4190. I therefore remember theirs as area code, exchange, and then moving the 1 left 2 digits... 3 "facts"...

[Terry McCracken]

In an experiment with Judith Polgar, they found (with a brainscan) that she actually uses the facial recognition part of the brain to recognize chess positions (/blocks).

Tony

Fascinating. I had suspected that might be the case, but had not heard it was confirmed. I seem to recall GMs can remember something like ten thousand or more chess patterns (AKA chunks). I wonder how many faces a single individual can recall.

It was done pretty funny. They redid some famous experiments. They had Judith sitting on a terras, drinking coffee, when a truck came by with the picture of a chessposition on it (1 by 1 m) . She saw it for about 3 secs.
5 minutes later she was asked if she remembered the position. No Problem.

Then she was told another truck would pass by within 2 minutes. It had a random position. She was unable (though she was warned and had more time) to setup the position afterwards.

Tony

I think that other GM with similiar rating to Judit polgar have better talent for chess and maybe they may do better in memorizing random positions.

I believe that Judit had more support to invest time in chess when she was a child relative to other 2700 GM's.

Uri

Randomised positions are far more difficult to remember as they don't make sense to the brain.

Organized positions do. That is why faces are simple to remember inter alia, is due to symmetry. Asymmetrical stuctures are more confusing to the brain and often the brain will ascribe patterns to these more nebulous shapes distorting what you really see, understand and remember.

An example would be seeing faces or animals etc. on walls or floors with randomised patterns, within clouds this phenomenon is even more pronounced.

Terry

There is a well-quantified "rule" about this... 7 +/- 2

The normal human can remember 7 random things, with an error margin of +/- 2. So some can only get 5, others can get 9, and I am sure there are a few way out on the standard deviation curve that can do somewhat better or worse.

The issue is "what are these things". In a random position, a piece/square is a single thing. In a normal position, the pawn structure might be one thing, the normal positions for pieces, etc. So for a GM the entire position might be a single thing if he has seen it enough.

For example, many of us remember 10 digit phone numbers as 6 pieces of data, the area code, the exchange, and then the final 4 digits are random.

In my case, my phone number is the same as the local auto-zone except the last 4 digits of my number are 4901, while autozone is 4190. I therefore remember theirs as area code, exchange, and then moving the 1 left 2 digits... 3 "facts"...

I'm well aware of this, and I don't believe I was arguing to the contrary.

[M ANSARI]

Chess up to a level of around 2500 ELO can be taught to almost anyone with normal intelligence and is fully dedicated to that task from a young age. Now to get to the upper levels you need something else, usually that comes in ability to memorize positions more than the average person. A strong GM can sense danger much earlier than others without actually calculating where the danger or opportunity is. He will simply play a strong move that will only make sense many moves later. I have a feeling that we have yet to see a truly strong super GM ... and when we do ... he will most likely be a savant that has many mental deficiencies in other parts of his brain. I watched a Discovery channel program on savants the other day ... in one case there was a young person that was able to look at a vista or a building and then be able to very accurately draw it with incredible detail .... this person had many mental deficiencies in other areas. As this person grew up, he improved in his mental deficiencies but that unfortunately caused his drawing ability to disappear. Someday we may be able to harness our brain similar to the way savants harness their brains in very narrow sectors without the negative side effects ... but for the time being I like my brain just the way it is thank you.

[Uri Blass]

Chess up to a level of around 2500 ELO can be taught to almost anyone with normal intelligence and is fully dedicated to that task from a young age.

I do not claim that you are wrong but
I wonder if there is an evidence for this theory.

Many people believe that you need to have special talent to get to the level of around 2500 elo.

It may be interesting to see what happens if some rich person choose 100 childs (6 years old) by random and give 1,000,000$ for everyone who get rating of 2500 in the next 20 years(the rich person should also give less money fpr smaller achievements of the same players so they are not going to give up because they do not believe that they can get 2500 in the first place)

possible prizes should be something like the following:

10,000$ 1400
20,000$ 1500
50,000$ 1600
100,000$ 1700
150,000$ 1800
200,000$ 1900
250,000$ 2000
300,000$ 2100
400,000$ 2200
500,000$ 2300
700,000$ 2400
1000,000$ 2500

How many of them are going to achieve 2500 and get 1,000,000$?

Uri

[Roman Hartmann]

Laszlo Polgar kind of proved the point of M Ansari already. His 3 daughters were trained in chess from early childhood and all of them were/are strong chess players with Judit beeing the strongest of the three.

Roman