Newsgroups: comp.ai
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!uupsi!cmcl2!acf5!hpd8626
From: hpd8626@acf5.NYU.EDU (Hasns P Dettmar)
Subject: Re: Another chess question
Message-ID: <12450002@acf5.NYU.EDU>
Sender: notes@cmcl2.nyu.edu (Notes Person)
Nntp-Posting-Host: acf5.nyu.edu
Organization: New York University
References: <1991Mar26.162003.7849@swift.cs.tcd.ie>
Date: 5 Apr 91 22:44 EST
Lines: 42

/* acf5:comp.ai / jeclarke@swift.cs.tcd.ie / 11:20 am  Mar 26, 1991 */

	I remember reading somewhere (possibly years ago) that chess
	Grandmasters did not gain their advantage over lesser players
	by being able to look more moves ahead than them (I think they
	used only look 3 or 4 moves ahead), but by the fact that they
	could eliminate all the "useless" moves from any position, and
	so have more time to concentrate on possibly useful moves.

	I seem to remember that this process was so unconcious that when
	these useless moves were pointed out to them they didn't reply
	that they had seen them and realised that they were no good, but
	that those moves had never even occured to them.
	
	The idea is that the expert chess player has built up a knowledge 
 	base of about 50,000 positions.  Each position in turn has associated
	with it semantic, visual, strategic, and tactical information.  This
	information then allows the player to make a "gross" judgment of the
	position, thus eliminating certain useless moves.       

	I did some research a couple of years ago with some grad students at
	NYU dealing with the nature of chess expertise.  I tested "expert"
	chess players and your basic joe schmo chess player and asked each
	to memorize a position on a board which was shown for only a few secs.
	Sometimes the user had to verbally say a completely unrelated 
	sentence, thus providing an interuption.  After this the players were
	asked to reconstruct the position.  My hypothesis was that due to the
	nature of the knowledge base, the experts would be less affected.  But 
	what I found was the opposite: experts performed significantly worse
	when there was some interuption.  Novices performed at about the same
	level.  What does this mean? One idea is that the knowledge base was
	so sensitive that the interuption completely screwed the expert up.  
	On the other hand, the novice doesn't have the knowledge base built
	up the point where it becomes a basis for expertise, so an interuption
	doesn't influenc him/her as much 
	Does this ring a bell with anyone? If so, I'd really appreciate
	the reference.

	Thanks in advance.

	John
/* ---------- */