Columns | September 11, 2012 14:49

Thinking and looking

Daniel Kahneman’s Thinking, Fast and Slow

Although it’s not the most popular genre around, there is definitely a market for chess books on psychology and subjective experience. For me, Jonathan Rowson’s Chess for Zebras (2005) is still one of the highlights in this category. While reading this book, I often found myself wondering how I had managed to survive without it!

I had the same feeling when I recently read Daniel Kahneman’s fantastic book Thinking, Fast and Slow (2010). In this book - which despite its appealing title is not a chess book but a book on economy and psychology - Nobel-laureate Kahneman explains the way we think. He describes how our thinking can be divided into two ‘systems’:

  • System 1 operates automatically and quickly, with little or no effort and no sense of voluntary control.
  • System 2 allocates attention to the effortful mental activities that demand it, including complex computations. The operations of System 2 are often associated with subjective experience of agency, choice and concentration.

Kahneman, who in 2002 received the Nobel Prize in Economic Sciences, argues convincingly, with vivid examples and surprising experiments, but I was already convinced after reading the book’s Introduction, in which he outlines the two systems. To a chess player, the idea that there is an intuitive and a rational side to thinking feels immediately right.

What made the book so fascinating for me weren’t so much the many experiments, theories and conclusions (although these are mind-blowing in their own right), but rather the aspects of psychology and economic thinking that Kahneman discusses in the process.

Take his chapter on the ‘Endowment Effect’, a basic economic concept on goods that are not traded regularly. In this chapter, Kahneman also mentions indifference as an important driver of economic thinking. (It makes a great difference whether people care for a specific amount of money to be spent or not; it depends, for instance, on whether they have a good salary or a lousy one!)

I’ve never read anything on the concept of indifference in the context of chess before, but I instinctively felt that this is a very important aspect not only of economic theory, but also of chess psychology. I immediately thought of those instances where my opponent’s apparent indifference towards the game - and its potential result - profoundly influenced my mindset and concentration to the point of embarrassment.

For me, it’s obvious that any opponent who clearly cares less about the result than I do - or thinks he already knows the result because he has a superior rating - has a psychological edge on me even before the game has started. This even works without actually seeing the opponent: on FICS and ICC, there are thousands of so-called ‘clickers’ (players who play hundreds of games a day without seeming to care or worry about the quality of their play) whose blatantly indifferent style of play is deeply unsettling to serious-minded people like myself.

As said, I haven’t read much about this subject in “real” chess literature. But why? Perhaps it has to do with the fact that it’s usually strong players that often seem so utterly indifferent when playing chess against weaker players – and chess books are often written by these very same strong players!

This “indifferent GM”-phenomenon was recently confirmed once again when a team member of mine confessed to me that he really felt demoralized (and irritated) while his opponent, a promising young Dutch Grandmaster, looked utterly bored during the entire game (which he duly won). He couldn’t help wondering why this youngster spent his time playing chess at all – wasn’t he better off chasing girls rather than his opponent’s pieces?

The interesting thing is that this indifference (which often turns into a distinct air of superiority) not only shows during an actual game (when it could be justified by the strong play the GM is showing), but also before and after, when there is no such thing as a game of chess going on at all.

This is strange, when you think about it. Apparently, the stronger player senses that he can afford to act in an indifferent or superior way, even though he is probably dealing with someone he’s never seen before. (The universal moral picking order in, for instance, tournament press rooms, is a fascinating thing to observe.)

Most weak players instinctively accept this status quo like it’s the natural order of things. In post mortem analysis between strong grandmasters, there is a tacit agreement that weak players do not interfere or ask ‘stupid’ questions, even though we now know, thanks to computer engines, how often these analyses contain serious gaps or flaws. But despite these tactical errors, we all know that the grandmaster ‘grasps’ the position infinitely better than we do. This ‘expert intuition’ is the Holy Grail in chess.

Kahneman discusses this phenomenon as well in his book. When I read the title of his chapter, ‘Expert Intuition: When can we trust it?’ I was reminded of those numerous times when, in post mortem game analysis among friends, someone started a discussion (not to say a row) by confidently declaring, “I may not be able to prove it concretely, but I still feel intuitively this has to be the right move!” I must admit that ever since the rise of computer chess, I’ve become more and more skeptical of this type of argument, even when I hear grandmasters proclaim it. It just happens too often that their feeling was, in fact, wrong.

Still, it’s undeniably true that many strong players have a fantastic ‘feel’ for positions and that their intuition is sometimes proven right only after many months of deep computer analysis. Indeed, Kahneman acknowledges this for chess specifically:

"If subjective confidence is not to be trusted, how can we evaluate the probable validity of an intuitive judgement? When do judgments reflect true expertise? When do they display an illusion of validity? The answer comes from the two basic conditions for acquiring a skill:

  • An environment that is sufficiently regular to be predictable
  • An opportunity to learn these regularities through prolonged practice

When both these conditions are satisfied, intuitions are likely to be skilled. Chess is an extreme example of a regular environment, but bridge and poker also provide robust statistical regularities that can support skill. Physicians, nurses, athletes, and firefighters also face complex but fundamentally orderly situations. (…) In contrast, stock pickers and political scientists who make long-term forecasts operate in a zero-validity environment. Their failures reflect the basic unpredictability of the events that they try to forecast."

Here I must disagree slightly with the great scientist. I think there are many sufficiently irregular situations in chess to question the universality of Kahneman’s statement. Intuitive judgement in a razor-sharp Najdorf without computer assistance will not hold up for long, as any grandmaster who has studied this opening will acknowledge –  and any grandmaster claiming the opposite should be challenged!  It is often said of Fischer Random chess, where there is a high degree of irregularity especially in the opening phase, that it benefits ‘intuitive’ players. Again, reading Kahneman’s conclusions, one might start questioning this.

I could go on and tell you about many more fascinating examples from Thinking, Fast and Slow, but the point I want to make is merely that one often reads more interesting things about chess in books on psychology or economy than in books about chess itself. Rowson is in fact one of a very small number of authors who do go beyond your usual A.D. de Groot reference.

This is odd, and also a bit scary. It shows, in my opinion, how little most chess authors think outside their own field of expertise. And how much there still is to learn and understand about the way we think in chess - if only we looked around us a bit more from time to time.

Arne Moll's picture
Author: Arne Moll

Chess.com

Comments

Henk de Jager's picture

Nice review. Another, new "out of the box thinker" in chess writing may be Willy Hendriks, his Move First, Think Later in a funny and intelligent way questions the way we are taught to think about chess and how we might do better.

Brian Karen's picture

IM Arne Moll column is consistently thought provoking. I believe he has hit on an important point. Apathy can be a big advantage in chess :).

Often players who get near their rating goal will find it hard to break the barrier the first time around. They start to care too much about the result as it becomes closer. But eventually it becomes routine to get to that level and they overcome the hurdle.

I recall when I first started at chess I would play my computer all the time. They were still beatable back then (late 80s). The first few times I blundered and couldn't crack the level. But once I started to routinely outplay the computer the result became less meaningful and I wouldn't 'choke'.

Activity at chess should lessen the pressure. A player who plays less frequently will care more about the result of each game because he will have less of a chance to 'fix' things when his rating drops. If he played more often the result of each game is less significant.

Brian Karen's picture

IM Arne Moll column is consistently thought provoking. I believe he has hit on an important point. Apathy can be a big advantage in chess :).

Often players who get near their rating goal will find it hard to break the barrier the first time around. They start to care too much about the result as it becomes closer. But eventually it becomes routine to get to that level and they overcome the hurdle.

I recall when I first started at chess I would play my computer all the time. They were still beatable back then (late 80s). The first few times I blundered and couldn't crack the level. But once I started to routinely outplay the computer the result became less meaningful and I wouldn't 'choke'.

Activity at chess should lessen the pressure. A player who plays less frequently will care more about the result of each game because he will have less of a chance to 'fix' things when his rating drops. If he played more often the result of each game is less significant.

Niima's picture

Fair enough, but then you also the argument that to improve your results in the long-run, you need to set time aside to analyse your losses (from chess and psychological viewpoints), which takes time and prevents you from playing too often.
I do not think that being in a relaxed sate and a careless one are the same thing. There is a way to concentrate during a game (which is necessary for avoiding errors) without getting too tense (which causes tunnel vision). It is a fine balance and hard to achieve.

Joe Lux's picture

You are disturbed by your higher-rated opponent's indifference? You can never improve that way. I relish the opponent's indifference!! That is the only way you can start defeating masters regularly: the ultimate goal. Work hard during the game. (If you did, you will remember every move for a week.) Think to yourself, "Oh, great chess master, teach me why my move doesn't defeat you!" They won't always do it.

Luca B's picture

Actually the book shows how we are guided or misguided by our system 1 during everyday life, while we are convinced to act according to system 2. Thsi may not be always true in chess, but shows how our brain looks for semplification thus causinmg many damages and misjudgements. The book is especially useful for chess players and traders on financial markets. Available heuristics is maybe the principal semplification process and it's one of the main patterns used by anti-cheating algos in online chess portals (semplification vs complication is often the difference btw human and engine)

Luca B's picture

It's interesting to see how even in the new book Lesson with a GM 2, the author mentions Kahnemann's book inthe epilogue.

ablos's picture

GM fontaine called Wang Yue panda kungfu, the commentator from france mocked and insulted the chinnese player during the 2012 chess oylmpiad in turkey. see video 11 in chesstv.com

Amos's picture

"For me, it’s obvious that any opponent who clearly cares less about the result than I do - or thinks he already knows the result because he has a superior rating - has a psychological edge on me even before the game has started."

Now, observe Magnus Carlsen during a game. He so often seems bored by it. Maybe this gives him edge over his GM opponents?

pundit's picture

Nah, because throughout his career Kasparov had just the opposite demeanor as Carlsen. So I don't think either demeanor is the reason for the dominance of either player.

Wim's picture

Brilliant book!

Henk de Jager's picture

This may explain Kasparov´s relatively mediocre results against Kramnik: he simply thought too highly of him.

Adolfo's picture

The problem with “intuition”, both in chess and in life is that it reflects a concept that we all know what the user means, but we also know (even the user) that such thing doesn’t quite exist. In other words, is useful and short; it is more precise and convenient to call a decision (chess or life one) intuitive, that to explain that it was made on general grounds, maybe with some element of knowledge and experience, however not very thought trough or well proven. In any event, we could never rely on it and could easily (as easily as it was made) be wrong, partially right or even right, according to many factors, including of course luck.
As for Dvoretsky classification of players in “Intuitive” (Karpov, Capablanca, etc) vs. “Logic” (Kasparov, Botvinnik, etc), as opposed to the traditional “Positional vs. tactical”, again, I would call it right idea; wrong execution (rather wrong word). Maybe “Abstract thinkers” vs. “Concrete thinkers”, “Inductive” vs. “Deductive”, you name them.
Finally, as far as having a higher rated across, with an apparent dismissive attitude towards the course of the game:
a) I could care less about any moron online, even Naka, playing a lot of garbage, sacrificing it all only to win or lose followed by some “LOL” comment. Same for those offering draws, never resigning, crazy flagging, or doing all of that (too much to name); screw them. My advice is simple, I take even online blitz as a learning opportunity, with those opponents that opportunity is denied: find a worthy one.
b) If it is in a classical timed rated game OTB, maybe that GM game was indeed bored (your friend had a bad Bishop vs. a good GM Knight and that was to be won in 50 moves), maybe the kid was just too arrogant, or maybe he is simply one of those – I´ve played and known some of these- with a complete relaxed attitude OTB ; I can think of Carlsen, Kramnik and some other unknowns to name, opposed to that freaky Kasparov, Naka “ I am driving crazy” face-body manner way, shown at least in critical moments. Believe me, as far as being annoyed or intimidated, most players would feel so vs. the latter group. I am not a psychologist, or even an economist (tough I do study that), but I tend to think that if a practical psycho- algorithm to use in each case should exist, there should be nothing more practical to learn (and practice) how to focus on the chess that is going on the board (remember who said “I don’t believe in psychology; I believe in good moves”), and don’t care about who is in front or what is he doing. I end with a small personal anecdote: The other day I was playing blitz online vs. some “JoeyG” on Playchess.com (not much higher rated than me there): he seemed (as far as his chess went) to be acquainted quite a bit better with his openings and to some extent with his chess handling than me: though for some 5 games the score was some +1 to him (+3;-2). Suddenly, I finger him, and I realized I had been playing vs. one of my “childhood” heroes, Joe Gallagher (his KG and KID books were among the first and most appreciated books I’ve ever read). I freak out a little; and started to lose more, even from 2 winning positions. Later, I figure, WTF, let´s play again as I was at the beginning, let´s play the board again, as if I was playing some patzer, or FM or whatever, why would I care; and quickly the course of the games leveled again. Conclusion: We should learn not to care who the opponent is, or what he does; that is easier said than done, but I believe is achievable; to quote Rowson (off the top of my head, apologies in advance for some inexactitude) as Moll did here “ When I am in good shape, “hope” plays no role in my play”.

JR's picture

I agree with Arne's statement. Being a 2000-2100 player myself, I have had the opportunity to play against quite a few titled players (albeit in rapid games mostly). In my experience, I have a bad score against IM's and GM's who didn't seem to be interested in the game at all (which can indeed be quite annoying). Such games happen to take place in the first rounds of tournaments.

Also, I score relatively better against titled players who seem to care more about the result. A part of the explanation seems to me that these games are played in a later stage in the tournament and I am in good shape (otherwise I wouldn't play them in that stage of the tournament), but I also feel that the 'indifference argument' has a certain value in that cases.

Moreover, reflecting on my best results and my most memorable games (for me personally, of course), I find that I usually felt indifferent about the result of those games and was just enjoying them and the problems that arose.

Regardless of the way indifference exists due to (a 'healthy' amount of) arrogance or joy in the game, the argument that indifference improves results makes a lot of sense to me.

rpxcdu's picture

quprckn

David's picture

Kahneman's book was excellent. I read it completely independent of my interest in chess as it's in another area of interest to me.

Some seem to have misinterpreted what "intuition" refers to here. Intuition in this case only means the intuition of an experienced expert, and this distinction makes all the difference in the world. The idea is that oftentimes in complex situations solutions cannot be worked out with logic in a step-by-step manner. In these cases, an expert uses his experience and expertise to "intuit" the path to a solution. The degree of success is based on his level of expertise, of course.

The basic idea of the two types of thinking- fast and slow- are these:

Fast thinking is "automatic" thinking: Driving a car (for an experience driver). Riding a bicycle. Very little conscious thought is required to do these well for most people, even though these tasks are actually extrordinarily complex if you think about them.

Slow thinking is when one must stop and consciously do mental work: solving moderately complex math problems, for example. Mentally assessing the outcome of a particular line in a chess game.

The way these might work together in a chess game, is that the expert will intuit ("fast thinking") in a given position which lines are worth considering, then use "slow thinking" to assess those lines. In the case of chess, this makes an enormous difference, as the non-expert will spend a lot of time just trying to figure out which lines might be worth calculating.

Latest articles