Reports | January 22, 2010 18:42

Jonah Lehrer on Carlsen and chess intuition

CarlsenJonah Lehrer, author of Proust Was A Neuroscientist and the recent How We Decide, has an interesting blog post on the recent TIME profile of Magnus Carlsen (about which we also reported). According to Lehrer, the use of computers actually enhances chess intuition.

In his post at The Frontal Cortex, Lehrer argues that (chess) intuition is not something 'semi-mystical', but consists of 'experience embedded in the unconscious', a point several others (such as Malcolm Gladwell in his best-selling book Blink) have made as well. Lehrer observes:

One of the fascinating elements of Carlsen's talent is that he's learned the game by playing computer chess, matching his wits against advanced algorithms. The end result is a prodigy who's amassed an unprecedented amount of deliberate practice at an early age (...).

At first glance, there is something surprising about a teenager weaned on chess software extolling the wonders of intuition. It's as if we expect Carlsen to act like his software, to be as explicit in his strategic decisions as Deep Blue, the IBM supercomputer. But that misses the real purpose of practice and the real genius of the human brain. (...)

When experts evaluate a situation, they don't systematically compare all the available options or consciously analyze the relevant information. Carlsen, for instance, doesn't compute the probabilities of winning if he moves his rook to the left rather than the right. Instead, experts naturally depend on the emotions generated by their experience. Their prediction errors - all those mistakes they made in the past - have been translated into useful knowledge, which allows them to tap into a set of accurate feelings they can't begin to explain.

Lehrer concludes that this should make us less surprised about the fact that 'a chess prodigy raised on chess computer programs would be even more intuitive than traditional grandmasters.' Of course, just how intuitive traditional grandmasters were is still pretty unclear. For instance, in the book Secrets of Chess Intuition (2002) the authors Mikhalchisin and Beliavsky devoted much time to Mikhail Tal, possibly the most intuitive player in chess history, who definitely didn't own a computer. Was Tal really less intuitive than Carlsen? One commenter reacted to Lehrer's last statement with reserved skepticism:

That is quite a leap. The truth is we have zero idea why this young genius is more intuitive, nor do we even have any way of knowing if he is "more intuitive" since it's such a subjective and abstract thing to measure.

This sounds like a reasonable argument, though I think Lehrer may be onto something all the same. Another prodigy playing in Corus – Wesley So – has often said that he, too, has mainly relied on computers when learning chess, and I'm sure more examples can be found. Indeed, when you look at how many young kids are true experts at handling computers when it comes to playing games, typing, or even truly complex stuff such as object-oriented programming, it sounds plausible that computers can enhance (chess) intuition in many ways.

Still, mere exposure to more positions is not the same as being able to use them in a sensible way. The idea that working with computers can influence intuition is similar to the idea that the internet may be changing the way we think.  This very question was also the subject of this year's Edge Annual Question, on which neuroscientist William Calvin answered:

Assembling a new combination ("associations") may be relatively easy. The problem is whether the parts hang together, whether they cohere. We get a nightly reminder of an incoherent thought process from our dreams, which are full of people, places, and occasions that do not hang together very well. Awake, an incoherent collection is what we often start with, with the mind's back office shaping it up into the coherent version that we finally become aware of — and occasionally speak aloud. Without such intellectual constructs, there is, William James said a century ago, only "a bloomin' buzzin' confusion."

It seems to me this is the real secret of these prodigies: how can they assemble all the information that enters their head into something as coherent as winning a chess game against a world class opponent? Until we have an answer to this question, it's perhaps not so strange people still call it a mystery.

Share |
Arne Moll's picture
Author: Arne Moll

Once an ambitious chess youngster studying Russian literature, Arne now is a "semi-retired" chess enthusiast working in the banking industry, writing columns and book reviews for ChessVibes in his spare time.

SmartChess! - Revolutionize your game

World Youth Under 16 Chess Olympiad

Comments

Rothschild's picture

A very interesting article. All this criticism of the term "intuition" is meaningless when presented in such a hostile and ignorant manner. Intuition is a term used in LOADS of research in cognitive psychology. Intuition will be greater with greater experience (number of games played/positions evaluated) as procedural knowledge accumulates. Read Damasio "Descartes' Error" for more on how this intuition works in the brain and makes us so superior to any machine in complex tasks.

Meppie's picture

Great article.
When I finished reading, I thought about something else, not mentioned here.
Nowaday a young chessplayer can at a very young age practice with a strong trainer/opponent (a computer that can beat the world champion). In the passed talents started practicing with a strong person when they already had a certain strenght. This must have influence on there develepment and there intuition.

Philipp's picture

I find the opinions of Lehrer are rather unsubstantial.
Why does he single out Carlsen? And Carlsen has indeed exclusively learned chess by playing a computer?
What does intuition mean?
This article is non-scientific rubbish

Adrian's picture

Magnus have commented himself that the biggest difference between him and Kasparov is that Kasparov is calculating, while he relies more on his intuition when playing chess

Alexander's picture

I have to agree with Phillip. The term "intuition" is far from defined. In fact, when people talk about intuition in chess, they usually appeal to your very intuitive understanding of the word "intuition", rather than conceptual.

But there nevertheless is consistency of the use of the term. We talk about "intuitive chess move", "intuitive style of play" and "intuitive sacrifice" when we are faced with chess phenomenons that are seemingly impossible to consciously calculate with mere human resources. Whenever we see sacrifice with enormous complications, or an subtle positional move whose meaning will disclose twenty plies later, we say that a player relied on his intuition. In most of the cases, this is tantamount to saying: "He didn't know what was he/she doing. He/she just did it."

Intuition in chess therefore denotes all mental processes that are not consciously present to the player but nevertheless have a decisive effect on his play. This is the way we tend to use the term of intuition when discussing chess.

But the moment we try to construct a scientific (or a philosophical for that matter) concept out of this common usage, we are doomed to failure. The problem is in my opinion the following one: not all the unconscious mental processes that condition our chess performance are chess-related. For example: if before an important game a chess player fulfills a certain routine he values as important for his performance (like Kasparov's famous pre-game lunches, or Kramnik's swims and walks), he will approach the play more confidently. With the routine accomplished, he will likely execute a risky sacrifice that he would otherwise pass. Or another example: a man who has just been on a date with a lovely girl, or a player who had had great successes in business, may very well play a game of chess with a confidence he got from those to chess unrelated sources. This is a phenomenon I think we all are familiar with: we play better chess in a cheerful or self-confident mood. We trust our judgment better; we feel we don't need to consciously calculate every sub-variaton of a move we choose.

But this confidence-conditioning is of course subconscious. I have no doubts that Kasparov would laugh at a psychologist who would suggest that he have played a daring sacrifice because he *knew* he had had salmon for lunch. After his match with Karpov, Kortchnoi would too find it strange that his *knowledge* of a parapsychologist being present in the playing hall have somehow affected his play. This is the main paradox of the confidence-boosts (and decreases): even though they are necessarily based on the things we are aware of, they always work subconsciously. Even though Kasparov knew he has executed his favorite routine before the game, his confidence in himself and consequently his performance will increase only in the case he will believe that his lunch affect his performance on solely *physical* basis. His self-confidence will effectively go up only in the case that he doesn't know that the main function of his routine is exactly boosting his confidence. In a nutshell, confidence-boosters work pretty much as a placebo does. Their causality must necessarily remain unacknowledged.

The problem is then the following one: we've said that intuition in chess stands for all those mental processes that subconsciously affect our chess performance. But then Kasparov's knowledge that he had had a salmon for lunch effectively is a chess intuition, as funny as it seems. And similar, a strong feeling of confidence that remains after a successful date or a business meeting also counts as a chess intuition: a bold sacrifice played under the spell of that confidence will be an "intuitive" one.

But this is clearly absurd. Memories of a fish on a plate cannot possibly constitute a chess intuition. To our definition of chess intuition we must add another criterion: chess intuition denotes all *chess-related* mental processes that subconsciously affect our chess play. But exactly here the true problem rears its ugly head: because these processes are per definitionem subconscious, we cannot (neither we nor the player himself) effectively discern the chess-related ones from the chess-unrelated. Because the reasons behind "intuitive moves" are subconscious, they are in principle unaccessible to us.

So, to sum up this rather lengthy argumentation: whenever we qualify a move or a style of play as intuitive, we are in the last instance saying nothing other than: "It is impossible for us to determine the reasons behind this move; it is impossible even to know whether they had anything to do with chess altogether."

anatman's picture

Tal the most intuitive player in chess history?

Isn't the general consensus that Tal was able to pull off his sacrificial attacks because he was a superb calculator? Which I think is the opposite of relying on intuition.

chess's picture

in blitz games you mostly play inuitiv:) blitz gamers forget to calculate in normal games.

chess's picture

intuitiv

chess's picture

intuitively:)

Arne Moll's picture

@Philipp.
It's a blog post, not a peer-reviewed publication, giving Lehrer rather more freedom I'd say. He obviously singles out Carlsen because he's a great example and he's the one being discussed in TIME.
As for a definition of intuition, the whole point is that it's such an intangible concept, somehow 'felt' by most people but pretty hard to define. That's what, in my opinion, makes the discussion and Lehrer's point of view so interesting.

merlin's picture

I think there is some confusion here. Intuition helps on the evaluation of the position. A better understanding of how intuition helps, would come out by comparing "intuition" and "principles". If you try to find the best move in the position according to "principles" like "now i got to fight for the center so i have to play this" or "i got to keep the bishop pair" and so on, it is most certain that you will get confused and play the wrong move.Not because these principles are wrong but because there are so many of them and it is a very subtle case to understand which principle to follow. Intuition somehow manages to "show" you the right principle by the process described in the article.
A very nice question though is why players with experience, like korchnoi are not in the current top list of players.IMO here comes calculation.Calculation has nothing to do with intuition since when calculating you think moves in the most logical way like "now i am threatening this so he play that" and so on.Intuition is more a tool in evaluating a position but can't help you to see positions some moves ahead. So except the ability to "listen your Intuition", you got to have the ability to calculate as well.
Finally I think we got to declare Intuition from the word "mood" which i think it's the mistake alexander did. Mood may affect our confidence but DOES NOT affect our intuition . In fact it is the Intuition that comes to us and tell us to play more aggressively when we r in a good mood.But still the phrase "all mental processes that are not consciously present to the player but nevertheless have a decisive effect on his play" is NOT describing intuition but mood.Keep in mind that a bad mood doesn't let our intuition to work the way we would like to.

merlin's picture

My last paragraph wasn't so clear.What i wanted to say is that mood affect our intuition in the sense that it can let intuition do it's job or not .But surely does not change the content of intuition.

Alexander's picture

At Rotschild: As far as I'm concerned, I am not an opponent of the term. The concept of intuition is an old philosophical concept, usually denoting pre- or non- conceptual knowledge. It would be foolish to struggle against a philosophical tradition of which modern cognitive sciences are a mere sideline (and not the most fertile for that matter).
However, intuition in this particular context is in fact problematic. If we understand chess intuition as a special faculty or a special function of our brain performing outside of the scope of our awareness, then we run into trouble. If we want to keep the term meaningful, we have to discern between the moves that were executed on the basis of purely chess intuition and moves that were backed neither by conscious calculation neither by chess intuition, but by chess unrelated factors.
For example: a player may find a certain move plausible not because he subconsciously "knew" or "intuited" that it is good, but because the move appealed to him aesthetically. Now let's suppose that this player values the aesthetics of the game to such a degree that it becomes something of a second nature to him: in the moments of dense conflict, he subconsciously decides for a move he holds the most beautiful. The question emerges: would we call such a player an intuitive one? I think not, because even though he produces moves according to a subconscious or an unacknowledged criterion, the criterion (aesthetics) is not chess-related. But to establish whether such a criterion is in fact chess-related, the principle according to which he plays would have to become conscious, i.e. his play would effectively stopped being intuitive. This was the crux of my hostile and ignorant argumentation.

At Merlin: I agree mood was a misleading example. A better counter-example would be an aesthetic criterion of playing chess, which I use in response to Rotschild.
But nevertheless, before further debate, I think we have to distinguish between two functions that we usually assign to chess intuition: the first one is proposing a move and the second is providing a sufficient reason for its execution. Proposing, or finding a move in a given situation, is in fact always produced solely by our chess-related mental resources. Kasparov didn't find Rd4 because he was happy, or because he had eaten salmon. He found it because because of his pattern-recognizing skills, chess intuition, ability to recognize the essential, or some other faculty. Executing a move, however, can be assign both to chess related and unrelated considerations. He may have played Rd4 either because he have calculated it to the end, or because he liked the sight of it, or because his "chess intuition" told him so, or because he felt confident that day.
We see that chess intuition, as it is usually understood, explains two different stages of chess-thinking: firstly, it can provide an explanation of how a player *finds* a move, and secondly, it can function as a reason or a ground of its *execution*. I was of course referring only to second sense: it is absurd to claim that a player finds a move because he is confident, but not so when explaining why he had effectively played it. Only in this sense can a mood and "chess intuition" play the same function: both can serve as a ground of playing a bold sacrifice. If they are subconscious, they are, as I have argued, indistinguishable.

Again I've written too much, but let me add just one more thing: if intuition can really be enriched with experience, then it is nothing other than the ability of pattern recognition. But if this is true, then intuition does not mean the way you play chess, but the way you think about it; it is not about which moves and positions you execute, but which ones and how many you recognize and consider. In this sense, there is nothing paradoxical in calling a player who relies solely on his calculation an intuitive one, which again shows how problematic the term is ...

merlin's picture

@Alexander I think the problem is that you have never felt it and you r trying to understand it by purely mental processes (of course this is impossible). But the fact that you have never consciously understood it doesn't mean that have never experienced it (obviously) so i ll try to point it out to you. Before i begin i want to mention that intuition interfere to EVERY decision we make no matter what and people r different on their ability to hear it.

Now let's go to a point where a player has to make a critical decision .Let's say (for simplicity)that one choice could lead to positional game and the other to a more tactical one but still both of them would lead to equal game(let's leave aside for now the way he made those evaluations). The fact is that his UNCONSCIOUS would want to go to the kind of game he had more successes . But the unconscious is not the only thing that plays a role to this decision (obviously again ). His conscious is pretty important too. So if that man would like to practice in a tactical game or follow a specific principle he has in his mind he would make his decision based on that. Now let's talk about the aesthetic criterion.The fact is that someone that uses this criterion will have some experience based on this criterion and so based on that experience his intuition will enclude this criterion together with all his other criterions ( don't forget that even "playing positional moves" is a criterion, "moving your knight around" is a criterion, "keeping your bishops" is a criterion and so on ).But still the meaning of intuition should be crystal clear.It is the feeling we have about things in our subconscious . Keep in mind that intuition works even for decisions we make the first time based on similar decisions made in the past.

Now the phrase "if intuition can really be enriched with experience, then it is nothing other than the ability of pattern recognition" should immediately been deleted together with the entire paragraph.I hope that after reading my previous paragraph you can understand the reason.

Also i should mention that you should translate "chess intuition" to just "intuition".It has to do with decisions and that's about it...

Rothschild's picture

@Alexander: I fail to see why you think cognitive neuroscience or cognitive psychology is less fertile than say, philosophy. The man who wrote the article deals with intuition as a scientific subject from the cognitive science's standing ppoint, and philosophy is very important as a fundamental way of thinking about these concepts for every person indulging in this subject.

Now, you seem to define intuition as something that is without "concept". This is not precise in the field this article was written. The fact that we humans possess such a vast memory for patterns stored in memory and linked to feeling gives meaning to the fact that Carlsen says he sometimes can "feel" that a move is right. It has nothing to do with aesthetics, which is only muddying the waters with yet another concept. Feeling guides our actions in decision-making in so many ways (somatic marker-hypothesis) that to disregard intuition as something special (a special faculty, you called it) is to grossly overstate the power of our declarative capacity. So much of "rational" action we make is based on "intuition"/non-declarative knowledge in so much of what we do that it is actually not a very new idea presented in this article. Again: read Damasio who writes beautifully on the subject and gives very illustrative cases on people with brain damage impairing "intuition".

Rothschild's picture

@Alexander 2: When you speak of aesthetics as some sort of criterion that a player supposedly uses which makes his moves more about an explicit goal towards "beauty" and not intuition. This misses the point of the whole topic and the article being discussed. Take what is said in the article:

"When experts evaluate a situation, they don’t systematically compare all the available options or consciously analyze the relevant information. Carlsen, for instance, doesn’t compute the probabilities of winning if he moves his rook to the left rather than the right. Instead, experts naturally depend on the emotions generated by their experience. Their prediction errors – all those mistakes they made in the past – have been translated into useful knowledge, which allows them to tap into a set of accurate feelings they can’t begin to explain."

How this is linked to "aesthetics" or Kasparov's salmon meal is unclear to me. The whole point is that the accumulated experience in the form of patterns and complex principles is so great that, whereas a novice will be able to declare that he moved the rook to the left because that protects it and attacks another piece, an expert at MCs level will make moves based on such vast experience and understanding of patterns linked to emotional encoded memories of moves feeling "bad" or "good" that the reasons for making the move are far more complex than what he is able to easily declare (hence non-declarative). He can obviously say something about the obvious reasons for why the move is not horrendous, but beyond that, the reasons for making that particular move, will in complex situations some times be made on intuition which most certainly at the root be strongly linked to strong and valid concepts. Much of what is known from patients with types of cortical damage impairing the feeling of a right decision (intuition), is that they are inable to make ANY decision because they become like computers: analyze explicitly the pros and cons of every probable outcome of every available action and become paralyzed from making any action. In severe cases this can be seen in making the simplest of decisions, f.ex. selecting a proper birthday card for a sibling.

Your comment

Speak your mind

By posting a comment you are agreeing to abide our Terms & Conditions