Thursday, May 8, 2014

The Einstellung effect

The last decades big steps were made to better understand how the human brain works although at the same time we do realize that psychology is a science still in its infancy. Exactly this poor and incomplete information generates lots of distrust for psychology. Undeserved as even today there is a lot to gain by considering psychology. 

Since many years I have a big curiousity in better understanding the psyche. Maybe I even followed the wrong studies but at home there was no support for psychology so it became logically engineer. Anyway if today I notice an article or tv program about this subject then I won't ignore it and if combined with chess then it only strengthens my interest. Recently 2 nice articles on chessbase were published about the Einstellung effect: article 1 and article 2. In those articles is explained how earlier experiences often have an important influence in new choices which sometimes leads to strange errors.

In fact I already discussed some examples of this effect on my blog without using the scientific name. I am referring to my first article about chessintuition in which I showed that reproducing analyses not always gives better results than searching yourself the moves without any particular opening-knowledge. Knowledge is often incomplete or outdated. Our memory has limitations which leads that remembering things often partly succeeds or even completely goes wrong when variations are mixed. In my 2nd article about chessintuition part 2 I show the dangers of implementing earlier acquired patterns, themes because every small change in a certain position can be decisive if something works or does not work.

While these earlier articles are about the impact of experiences sometimes happened much before the game, this time I want to talk about the impact of experiences within a game which influence in a weird sense the further course of the game. In other words I want to demonstrate that the Einstellung effect can also appear in a very short timeframe. In fact this is nothing new as who doesn't know the next joke? Answer the following questions! Which color has the first white refrigerator? Which color has the second white refrigerator? Which color has the third white refrigerator? What does a cow drink?

Many people answer wrongly on the last question when they hear the joke for the first time as they are misguided by the earlier questions. I believe that I also encountered something similar in my interclubgame against the Bulgarian grandmaster Dejan Bojkov (which i earlier already fittingly mentioned in my article camouflage ).
[Event "Interclub Deurne - Amay"] [Date "2014"] [White "Brabo"] [Black "Bojkov, D."] [Result "*"] [ECO "C99"] [WhiteElo "2336"] [BlackElo "2500"] [SetUp "1"] [FEN "RR2nr1k/7p/1bqp2p1/4pp2/1P2P1P1/4N2P/3B1P2/5NK1 w - - 0 36"] [PlyCount "11"] {(I spent almost half of my remaining minutes to calculate the consequences of Ra6.)} 36. exf5 (36. Ra6 $2 {(I rightly choose for exf5 as winning the piece leads to fully unclear complications. )} Qxe4 37. Raxb6 f4 38. Ng2 Kg7 39. Rd8 Nf6 40. Rdxd6 Qf3 41. Nh2 Qxh3 $14 {(The engines show a slight advantage for white but it is all very unclear to me.) }) 36... d5 37. Nxd5 $4 {(Now easily winning was Ra6 as Qe4 is not possible anymore. Probably influenced by the earlier calculations, I forgot about the possibility and all my attention went to another combination which unfortunately was wrong. )} Qxd5 38. Bh6 Rg8 39. Bg5 $2 {(Only now I noticed that I did not have a winning combination but my move only make things worse.)} (39. Rxe8 {(Despite blacks answer this move is surely much stronger.)} Qxa8 {(I forgot completely this move.)} 40. Rxa8 Rxa8 41. fxg6 hxg6 42. Kg2 {(The endgame must be defendable for white.)}) 39... Qf3 40. Ne3 Bxe3 41. Bxe3 {(Here I was lucky to have Bg5 as answer to Nf6. About this endgame I likely will write another article as it is pretty interesting what happened afterwards.)} *
Because a move earlier after some reflection I rightly ignored Ra6, I wrongly didn't spend time later at it. The fact that I was in slight time-trouble and I believed to see a winning combination, will certainly also play a role but I consider the einstellung effect as the main culprit to miss the simple win. Now if the reader thinks "this can't happen with myself" then I do find it funny reading on the blog of (yes!) my opponent that he exactly wrote the same in the article the art of excuses. I quote: "It is funny how people (or it is probably just me) tend to miss recurring tactical themes in a single game."

The summit which I once experienced was a recurring tactical motif in a combination of which I completely missed the switch-back. Probably I wasn't very concentrated as I still can't believe how I could play such blunder after quite some thought.
[Event "Interclub Deurne - Aalter"] [Date "2011"] [Round "?"] [White "Brabo"] [Black "Draftian, A."] [Result "*"] [ECO "B41"] [WhiteElo "2325"] [BlackElo "2272"] [SetUp "1"] [FEN "r1b1k2r/2qp1pp1/pp2p3/4b2p/2PnP1P1/P1N1B3/1P1QBPP1/R4RK1 w kq - 0 14"] [PlyCount "8"] 14. Bxd4 $4 {(I totally miss the switch-back motif which black executes in the game. After e.g. f4 it is a normal game.)} (14. f4 $5 Nxe2 15. Nxe2 Bd6 16. g5 h4 17. Bd4 O-O $13) 14... Bh2 15. Kh1 Bf4 {(Only now I discovered the point.)} (15... hxg4 $2 16. g3 Bxg3 17. Kg2 {(I expected this in the game but even then Houdini finds an advantage for black.) } Bf4 18. Qd1 Rh2 19. Kg1 f5 20. Re1 Qd8 21. Bxg4 Qh4 22. Bxf5 Be5 $15) 16. Be3 {(In the game I gave up my queen with Bxg7 but kept on fighting a lost battle.)} hxg4 17. Kg1 Bh2 {(The missed switch-back by which black delivers mate in 2.)} *
Such blunders are not only painful but also very difficult to avoid. We are no engines which each position calculate from scratch and forget what happened before (with the exception of identical positions). It is often fascinating how engines in an instance can detect huge evaluation changes in the smallest difference of a position. Often I spend a lot of time to find out what exactly caused the Butterfly effect. I remember such recent case in my analyses of the game against the Dane Jens Frederiksen (a fragment was already covered in my article excelsior) in which engines pop up with the crazy move 9...,Kd7.
[Event "Open Leuven 2de ronde"] [Date "2013"] [White "Frederiksen, J."] [Black "Brabo"] [Result "*"] [ECO "C68"] [WhiteElo "1730"] [BlackElo "2347"] [SetUp "1"] [FEN "r3kbnr/1pp2ppp/p1p5/8/6b1/5N2/PPPPKPPP/RNB1R3 b kq - 0 9"] [PlyCount "10"] {(An interesting novelty recommended by the actual strong engines. )} 9... Kd7 $5 $146 (9... O-O-O $5 {(I chose for this standard-move which had been tested several times earlier in practice.)} 10. d3 Re8 11. Kf1 Rxe1 12. Nxe1 { (Blacks play is somewhat more enjoyable but white is still very solid.)}) 10. d3 Re8 11. Be3 {(Here we detect the difference with 0-0-0 as after Kf1 black can answer with Bxf3 because the rook at e8 is defended. )} f5 12. h3 Bh5 13. Kf1 Bxf3 14. gxf3 $13 { (White has to play very precise to avoid standing worse.)} *
To withdraw from the Einstellung effect is very difficult or maybe even impossible for most of us. I notice this also for myself as I do some quick unconscious actions which I don't fully understand afterwards. Probably resistance can be built up by forcing oneself to try out regularly new unknown stuff so by leaving the comfort-zone. Easier said than done of course as the temptation to win easy points based on known stuff is very big.

Brabo

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.