Random Posts

Sunday, February 23, 2014

Followup on Chess and Aging; also Learning and Calculating

     This is just some additional miscellaneous thoughts related to the previous post and some thoughts on the chess players' main  task, calculating variations.
     Two new studies suggest that aging may be kinder to chess champions and competitive Scrabble players. The first study, appearing in the June issue of Psychology and Aging by Florida State graduate student Roy Roring and Florida State psychologist Neil Charness, PhD, shows that top players' skills decline slower with age than those of less accomplished players. Another preliminary finding published in the Journal of Experimental Psychology: Applied found that Scrabble experts may also age more gracefully on some cognitive tasks (i.e. thinking, reasoning and remembering).
     That study, conducted by Claremont McKenna College psychologist Diane Halpern, PhD, and graduate student Jonathan Wai of Vanderbilt University, found that Scrabble playing taps cognitive skills that chess playing does not, including the need to speedily access verbal, visual-spatial and mathematic abilities. Both studies add to the growing literature on experts that psychologists hope will help them better understand how the mind works and ages.
     "By studying experts we can sometimes understand the scope and constraints on human cognition," says University of California, Davis psychologist Dean Keith Simonton, PhD, himself an expert on personal success. The chess study, for example, "applies state-of-the-art methods to a novel data source to answer an age-old question: Is aging more gentle or more cruel to the highly able?" he says.
     Using a database created by University of Newcastle, Australia of the ratings of 5,011 chess players over time, psychologist Robert W. Howard, PhD used a technique which allowed his team to examine players' skill development and analyze their trends. His analysis allowed him to observe how players grew over time and allowed him to see how different variables affected growth. He examined at which age chess players peak and how fast they decline after that. The study showed that competitive players top out around age 43 which is up to 10 years later than calculated by the study conducted in 1986 by Arpad Elo.
     Charness's finding supports mounting evidence for the theory that skills that rely on speed, such as sprinting and tennis, may peak earlier than skills where speed is less important, such as golf and chess. What's more, the best players showed the slowest rate of decline with age. Competitive Scrabble players may escape some of the ravages of aging as well, concludes Halpern. In several studies, she and Wai compared the cognitive skills of competitive Scrabble players (whose average age was 41) with those of a group of high-achieving college students with an average age of 19. The Scrabble players had better memory for shapes, words and letters as well as placement of words and letters on legitimate and transformed Scrabble boards. They also reacted more quickly on a task that required them to visualize what a piece of paper would look like after it was folded in a certain way. These results suggest that decades of intensive Scrabble playing may have positive effects on some cognitive abilities, say Halpern and Wai.
     This study's results also support researchers' findings in other expert domains. In particular, the study found that getting to expert level requires thousands of hours of practice over many years. In another study it was found that to build genius, your learning program must be based on high applicability of newly acquired skills and knowledge. If you memorize the whole phone book (i.e. a big set of facts), you won't be much closer to a genius mind and your problem solving ability will increase only slightly. To accomplish smart learning, you will need to constantly pay serious attention to what material you decide to study. You must avoid short term gratification at the cost of long-term learning. Some learning is fun and entertaining, but it has no real payoff. Your study must be made with an attempt to gain understanding. You cannot be guided just by the fun of learning but by your goals and needs. The term trivia excellently reflects the sort of knowledge we do not want to learn in the quest for genius. These are not-so-useful facts or rules of low applicability. 
     One interesting observation was that in the course of study, you may not be able to see or verbalize some rules but your brain will extract them in the course of practice. This helps explain why in many cases, strong players cannot explain how they visualize positions and come up with their moves; they just ‘know’ a move is good or bad.
     So, as chess players what should we study and how should we go about calculating? We all have problems calculating complicated positions. We can usually see a single line in considerable length but when the tree of variations becomes wide we find we can’t remember different positions and so our calculation’s suffers. The truth is it is not possible to calculate all variations in most middlegame positions. To do so is like the brute force method used by engines; we humans can’t do it.
     Move selection involves finding good moves, thinking ahead and judging the end position. Studies have shown that strong players instinctively choose stronger candidate moves than the rest of us. They don’t just stumble on good candidate moves; thier experience with similar situations (pattern recognition) and their knowledge suggests to them the right moves.
     When it comes to moving the pieces in your head there is only one thing you can do to improve this skill: practice! Another vital ability is the judgment of position. All your calculations, no matter how deeply you see into the position and how accurately you visualize it, are useless if your evaluation is wrong.
     One of the pieces of advice most often heard from strong players is that you should analyze master games. It has been suggested that many players don’t do this simply because most of the strategy, tactics and evaluations are beyond their comprehension. I would also suggest that another reason many players don't do it is because it is not fun. It involves giving up the short term gratification in order to gain understanding as mentioned above.  Gaining understanding usually involves thought and effort and many players are seeking an easy way to mastering the game. 
     For lower rated players it probably is best to avoid looking at master games with no textual explanations. There is no value in just looking at a ton of variations. We need things explained. It is always a good idea to cover up the next move and guess the next move. When I did this years ago I bought Informants and played over 600-1000 games and kept track of my score. It was dismal at first, under 20%, but after going over several hundred games it began improving…to over 70%.
     Pay attention to the opening principles and don't try to memorize variations. One thing I have noticed while playing a lot of online games is that even 1600’s violate opening principles with great frequency. When ‘anno-Fritzing’ my games afterwards more often than not it is my opponent who first deviates from book lines even when playing openings with which I am unfamiliar. It is also helpful to play over master games where the opening of your choice was played…pattern recognition.
     Go back over your games a week or two later. It’s surprising how much you will have missed on the first look! It’s OK to use an engine to look at tactical possibilities that were missed, but you won’t get much else out of using an engine to study games.
     When going over master games it’s probably best to stick to pre-1950s games and avoid contemporary games because the older games will be less esoteric and more clear cut. Some of my favorite game collections over the years have been: Tarrasch’e Best Games (Reinfeld), Reshevsky’s Best Games (Reshevsly), 100 Selected Games (Botvinnik), Zurich 1953 (both Bronstein’s and Najdorf’s book), Alekhine’s Best Games (vol. 3 by Alexander), My 60 Memorable Games (Fischer) and My 50 Years of Chess (Marshall)

No comments:

Post a Comment