IQ and Chess: The Real Relationship

This is a repost of an article that was lost from chess.com due to certain circumstances.

IQ and Chess – The Real Relationship

There have been and still are ongoing discussions on whether IQ is connected to chess or not. I shall, in this article, state my hypothesis on this matter and provide as much proof as I can. If there are any arguments against my case, I expect them to have some form of proof as I here provide.

Here is the breakdown of my article:

I. Hypothesis

II. IQ

  A. Is IQ equivalent to intelligence?

  B. The breakdown of IQ according to its developers

III. Chess

  A. How the brain works with chess

  B. Potential chess rating

IV. Chess and IQ

  A. Study of chess and IQ

  B. Which factors of IQ correlate with chess and the brain?

  C. IQ level and chess rating specifics

V. Likely Conclusion

I. Hypothesis

          There is a definite correlation between IQ (intelligence quotient) and general chess rating / level of chess play. This correlation can be attributed to several different factors which will be discussed shortly. The clearest correlation will be obvious in a particular chess rating range / group of players and with a particular IQ range.

 II. IQ

          IQ, or intelligence quotient, is a number that measures certain aspects that are considered directly relevant to intelligence. Researchers have spent many years attempting to prepare a test that can measure intelligence, but even the definition of true "intelligence" is debatable. Thus, there is no real measure of intelligence. A general intelligence factor (g) may be slightly more accurate in terms of measuring intelligence, but is still not close enough to measure overall "intelligence". Why? There are many different factors, including the physical state of a person; i.e. a blind person will score lower on an IQ test because he / she lacks the visual aspect required in the test; a person who hasn't learned to read because they cannot afford to, etc., may have the potential of being geniuses, but not the resources.

  A. Is IQ equivalent to intelligence?

          Absolutely not. As aforementioned, there is no truly accurate way to measure intelligence; IQ measures only particular aspects of a person's intelligence; aspects such as memorization of facts, talents in fields (such as art and music) which cannot always be measured as "good" or "bad" because of different tastes, and others are not included in an IQ test.

          In addition, it is possible to increase any given IQ with practice - since the IQ test focuses on certain aspects of intelligence, any person can do exercises that improve those aspects and leave all other aspects of intelligence untouched. Thus, someone may master the aspects in IQ tests, but still have no factual knowledge - is that person to be considered a genius? In general, that person would be considered a savant (amazingly talented in one aspect of life that requires powerful brain usage, but deficient in other activities due to a lack of brain power for those neglected aspects).

          Thus, one should never consider IQ to be a true, final measure of intelligence. It simply measures a relatively small given portion of intelligence.

  B. The breakdown of IQ according to its developers

          One of the original and most developed models of the IQ test is the Stanford-Binet model (Stanford revised Binet’s original version of this test to form this model of the test). The most modern version involves, generally, the following aspects:

          1. Knowledge (reasoning and applying facts)

          2. Fluid reasoning (problem-solving and patterns)

          3. Quantitative reasoning (mathematics)

          4. Visual-spatial processing (general patterns, seeing larger pictures)

          5. Working memory (used for short-term application of knowledge)

          6. Nonverbal IQ (includes patterns and other problems without words)

          7. Verbal IQ (comprehension, including vocabulary)

          Those are the main aspects tested in most IQ tests, to some extent. Generally, we could see that patterns, reasoning, problem-solving (including mathematical), and comprehension of words is necessary to master an IQ test. Is this not but a simplified SAT test plus some pattern recognition? And have we not been told that the SAT does not measure intelligence, but simply how well you will do in mathematics and English courses in the first year of college?

          Thus, the IQ test can be seen as a measure of the ability to comprehend English, mathematics, the ability to make connections when doing problems in the aforementioned subjects, and recognize patterns and apply an associated comprehension of them to complete the patterns.

          Now, we shall explore exactly what happens when you play chess – and we shall soon be able to connect IQ with chess, based on those processes.

NOTE: This part is heavily based on http://www.assess.nelson.com/pdf/sb5-asb1.pdf

III. Chess

          Chess is basically a battle on the board. What do you need in order to be able to play chess? You must be able to:

          1. See the board

          2. Know how the pieces move

          3. Correlate the movements of the pieces with each other

          4. Use the correlation of pieces to perform a certain objective

          5. Use a set of objectives to help you perform one ultimate objective (i.e. checkmate)

          Let us explore how the brain could possibly take on these seemingly incredible tasks.

  A. How the brain works with chess

          The initial task the brain encounters when first seeing chess pieces is the understanding of what those pieces are and what they do. When one first sees a chess board with pieces, there is utter confusion and the brain may signal the person to give up – but the person winds up associating these pieces with their respective positions on a board and how those pieces move.

          This is done when the eye registers signals about the positions of each piece and how the pieces are associated. The newly established visual signals are transferred to the brain, which attempts to make sense of the board and pieces via neuron connections. These connections use previous experiences and memories, seen as similar to those found a chessboard, to help a person comprehend the positions of pieces on the board. Thus, a person, subconsciously, is able to associate these pieces with each other. Next, to learn how the pieces move, a person must be able to connect these symbols / pieces to a particular change in position. Thus, the brain uses further neurons, already connected with the initial neuron connections used to understand the original positions of the pieces, to connect with memories and experiences that involved connecting symbols with functions (i.e. waving a hand signals a greeting or a farewell). Thus, the brain is able to connect the functions of the chess pieces with the pieces themselves. Finally, the brain connects the functions with the original positions of the pieces. Therefore, without actually paying attention to the symbols representing positions or changes in position, the brain is able to connect initial positions with changes in position. As a person learns how to change the positions of chess pieces, new neuron connections associate with the older neuron connections used for the original change in position for chess pieces. A person is therefore able to associate new positions with similar changes in position.

          Next, the brain must be able to associate the functions of the pieces with each other. This, of course, comes after the brain is able to master the positional changes of the pieces from any position on the board. At this point, visuals [of the board] are not necessary. The brain has already made enough connections to establish the visual-spatial relationships of the pieces and their associated functions; now, the brain can attempt to connect the functions of the pieces and focus them onto different positional spots established by memory. Thus, the brain is able to create combinations involving more than one piece in order to create an objective that focuses on particular positions on the board. This is how chess players ultimately train themselves – they begin to make more connections that allow them to logically see their combinations and responses to their combinations. This all done in the mind, and not actually on the board (at least not in OTB / over-the-board chess). It is obvious that playing out combinations directly on a board would make this process easier, but training directly within the mind, without the necessity of visual stimuli that represent the pieces, gives more difficult training that helps the neurons connect and allows for improvement of the game based on the connections of these neurons.

NOTE: The previous three paragraphs were based solely on my analysis of learning based on my knowledge of neuron connections. The true validity of this information is disputable, but nevertheless, my analysis should be roughly accurate.

          It is clear that, at some point, every professional chess player thinks deeply about moves within their minds. But is my claim that visualization of chess pieces is a “luxury” when thinking about moves and combinations true? You may want to test this out, and do so thus: visualize a chess board with all the pieces completely set up – then play several moves in your head. You should notice that your chessboard and pieces don’t seem to have a particular shape / color / size, or even dimensions that connect to how you see them on the chessboard. You seem to simply “know” where the pieces go and how they move without having to actually picture a knight or a King. This is extremely normal; in fact, an account of this experience has been described by a person who plays blind chess: http://www.notebook.kolchenko.com/intelligence/blind-chess-and-working-memory. I must note that the points I make in this paragraph on based on this person’s account and on my realization that I have nearly the same experience as this person whilst playing blind chess, or even when thinking about moves.

          Now, one may be able to understand how blind people are likewise able to play chess and improve at it. Once a person learns the pieces and how they move, the brain takes over and replaces the pieces with their directly associated movements / functions and connects with each other to achieve several goals: capturing pieces, gaining advantages, etc., to ultimately achieve checkmate.

          It should now be clear that initial visual stimuli are absolutely necessary for a person to be able to play chess. Next, it should be clear that visual stimuli are not necessary for a person to improve at chess. Finally, it is necessary to note that non-visual improvement cannot be achieved without initial visual stimuli.

  B. Potential chess rating

          This factor of chess has been widely discussed over many years, and may never be truly resolved. Nevertheless, it must be noted that, since the brain can always continue to make connections between neurons, rating can be close to infinity. It must also be noted that rating is only a comparison to other players, and, if seen only as a comparison, then rating is just a number. But let us see rating as a number that represents ability (which it does, to a reasonable extent).

          We must now consider that a chess rating of infinity can be reached if a person devotes his or her life to chess, practices nonstop, and has enough of an ability to connect previous chess experiences with new ones in order to improve to an infinite extent. But this is not the case – everybody can have limits to amount of connections they can make and to amount of effort they will put into chess!

          Since improvement in chess involves using short-term and long-term memory, and the knowledge of patterns and their associated connections, people may not be able to make extremely complex connections involving these patterns. In other words, after making an extremely large amount of connections, a person may find that they are unable to see beyond a certain point limited by their short-term visual-spatial memory. They can think ahead approximately 20 moves or so, and finally stop because any more thinking is far too difficult for them to fathom. The brain is unable to make any more connections without an effort that a majority of humans are unable to conceive. Thus, a person may reach their peak level of play and their maximum rating. Thus, they reach a rating that may be considered close to their potential rating.

          So, how should potential chess rating be defined? Potential rating would be the maximum rating a person can reach after making as many visual-spatial connections as he or she possibly can. After an incredible amount of connections, a person would be unable to progress without effort that could possibly cause the brain to overwork and thus create imbalances in short-term and long-term memory.

NOTE: the analysis of potential chess rating is based on my predictions based on my knowledge of neuron connections and the limits an average brain can achieve, based on factors directly related to chess improvement.

          Now that we understand what factors can contribute to a maximum / potential chess rating, we may be able to make a distinct connection between chess rating and IQ.

IV. Chess and IQ

          It should now be clarified that IQ involves several factors [which do not sum up to total intelligence] including: visual comprehension (of patterns, etc.), understanding of English to a limited extent, understanding of mathematics, and the ability to connect these factors to solve problems based on each of the three aforementioned areas. Another clarification, for chess rating: a limited ability to make further connections regarding combinations in chess can lead to a maximum or near-potential chess rating. Now, let’s comprehend a connection of chess and IQ based on previous studies.

  A. Study of chess and IQ

          A 1979-1983 study, described in an article by Dr. Peter Dauvergne in July 2000 focused on how chess affects the development of children. These children were made to study chess for a particular amount of time: 60-64 hours over a period of 32 weeks. Chess increased the children’s creative thinking significantly; creativity is not a factor of IQ tests, but some creativity may be important in recognizing complex patterns, sometimes seen on IQ tests.

          A 1980’s Venezuelan study of 4000 children, also described in Dauvergne’s article, showed a general increase in IQ (the IQ test used the Wechsler Intelligence Scale, which has a direct connection to the Stanford-Binet IQ test models). Thus, it is clear that chess has the ability to improve children’s intelligence, including creative (part of which is the visual-spatial factor, including pattern recognition), mathematical and even English factors. How could chess possibly improve English skills? The young students learned to make connections based on chess moves; as previously mentioned, further neuron connections could be made, based on previous memories and experiences (in this case chess experiences) to help students connect different aspects of what they read in English courses / texts. Thus, the ability to make connections improves the overall IQ score.

NOTE: the previous two paragraphs were based on http://www.auschess.org.au/articles/chessmind/htm

          Now we have understood, through a valid source, that chess can improve IQ. But since this is true, cannot consistent chess improvement cause IQ to indefinitely rise? Doubtfully so: first of all, we must consider that the subjects tested were children; children have a much greater potential to connect neurons, because they are yet relatively uneducated and have much more space to connect between neurons. They lack the experience to have made more than a few connections, and thus are able to learn (through these neuron connections) much quicker than adults! Children can consistently improve at chess and thus improve their IQ to a certain extent. At some point, their maximum ability to connect visual-spatial memories will decline. This is due to the increasing number of new problems and potential positions in chess (mentioned in Dauvergne’s article): at some point, the brain will only be able to connect previous memories to solve one problem, and then be unable to use the new memories to connect to a new, totally different chess problem. The brain will not see many similarities between the previous position and the current one (except in rare situations) because this position has been connected to a more basic memory in chess. Thus, a formerly created connection is reinforced, but new connections are limited.

NOTE: the previous paragraph consisted of my analysis based on Dauvergne’s article and my knowledge of neuron connections.

          Now that we have established a study of chess and understood why maximum ratings can truly be reached (even in children), let’s fully connect chess rating and IQ.

  B. Which factors of IQ correlate with chess and the brain?

          It should be quite clear that chess involves visual-spatial processing (as aforementioned several times in the article), some quantitative reasoning (i.e. recognizing the value of pieces, calculating some combinations that lead to advantage in terms of material), working memory (used to strengthen the brain’s ability for visual-spatial processing and comprehending current positions), fluid reasoning (recognizing and understanding patterns, especially of individual pieces and pierces working together), nonverbal IQ (further understanding patterns), and knowledge (applying previous experiences from long-term memory to the current situation presented in working memory / short-term memory).

          Thus, chess uses six out of seven factors of the modern IQ test model. It is obvious that not all possible combinations of quantitative reasoning, fluid reasoning, etc., used in chess are also used in IQ tests; nevertheless, chess does incorporate a significant enough amount of IQ test factors to have a direct connection to IQ. There is no clear measurement of exactly how connected chess is with IQ, but, upon seeing IQ test models, rough estimates can be made. If the brain does, indeed, use connections from IQ tests and from chess, there could be an incredible connection of 0.5 (50% connection). There is no direct basis for this estimate; it is, as previously mentioned, simply a very rough estimate based on the amount of factors that are connected between chess and IQ, and the relevance of chess and IQ factors and potential questions on IQ tests. The connection between chess and IQ can even be partially validated based on aforementioned studies showing that chess improves IQ.

          Now, we can deduce that improving our general strength in factors such as visual-spatial reasoning, quantitative reasoning, etc., could improve our chess abilities and thus our potential chess rating significantly. But our ability to improve chess rating and even IQ diminishes as the brain makes more and more connections and is eventually able to only strengthen previously made connections (as mentioned before). Thus, a more specific conclusion can be made to validate the real relationship between IQ and chess.

  C. IQ level and chess rating specifics

          It has been argued that maximum chess rating can be estimated through the equation Elo ~ (10 x IQ) + 1000 (which is considering intense study and practice of chess), according to the Levitt equation. An explanation of this equation can be found by Levitt himself on this website: http://www.jlevitt.dircon.co.uk/iq.htm. There is no direct basis to this equation, but according to the definite connection seen between chess and IQ, this could be a valid potential study.

          We must take that many people on this website have claimed that their IQ is high and chess rating low, or vice versa. But there is no proof that these people have yet reached their maximum or potential chess ratings or maximum or potential IQ’s, since both can be improved. It must be strongly noted that there may be a limit that a human can reach in chess rating and IQ if they put enough effort into achieving these feats. Only when these feats have accomplished and noted, can an utterly accurate study be done; with this study, it may be probable that chess rating will be proven directly related to IQ.

          Finally, we must comprehend that a majority of the improvement that takes place in chess occurs at the beginner’s level. Going from not comprehending even the most basic concepts of chess to making relatively simple combinations of several moves is an incredibly huge change, if you consider the differences. From not knowing chess to being able to perform these feats, people change from chess ratings of nearly 0 to ratings of possibly 1800+; this is far above half of the highest rating ever reached. Thus, it can be argued that a majority of IQ connections are used in the improvements of beginners.

          Now, we could look at the very basic learning of chess. Most connections for learning chess will be made when learning the positions and functions of the pieces; a large amount of connections will also be made when making simple combinations. Thus, a large amount of factors in IQ will be used for people rated between approximately 600 and 1400 (600 being the people learning the positions / functions of pieces and 1400’s being those learning and comprehending combinations of 2-3 moves or even slightly more).

          Next, we must understand that large differences in IQ will usually result in large differences in the abilities to master the several aspects of IQ tests (except in the cases of savants or other similar phenomena). Thus, a person with a maximum IQ of 60 will have much more trouble with chess than a person with a maximum IQ of 140. The person with the 60 IQ is more likely to lack the ability to combine the strengths of several aspects of the IQ test and connect them to chess. For example, let’s use simple mathematical formulas to calculate potential partitions of this person’s IQ (although this should never be done in professional IQ studies): using the six factors for chess, let’s partition 10 IQ points to every one of those six factors. Next, let’s partition 20 points to all seven factors (including the factor not used in chess) to the person with the IQ of 140. We can now compare the partitions: the person with the higher IQ has at least twice as many points in factors that involve chess as the person with the lower IQ. Thus, the person with the higher IQ immediately has a much higher potential to improve in chess because of the higher ability to perform in the aforementioned abilities. The person with the lower IQ has a much lower potential because of the more limited abilities.

          Now, we must once again take note that IQ can demonstrate a person’s potential ability to master chess; in addition, we must note that the subjects I have given have their likely maximum IQ’s determined. The person with an IQ of 60 may defeat the person of IQ 140 if the person with the lower IQ has played chess consistently over many years and the person with the higher IQ has just begun learning chess. Nevertheless, given equal effort and time, the person with the higher IQ will generally be able to overpower the person of lower IQ at chess.

          Having said this, it is clear that differences of IQ in the range of approximately 60 to 140 will give a fairly large margin for fair differentiation of chess ratings and abilities. In addition, a fair range of chess rating (600 to 1400) should be decent enough to monitor the relationship between general IQ and general chess rating / ability and associated progression.

          In other words, the largest, most accurate connections between chess [rating] and IQ will be seen with an IQ range of 60 to 140 and chess rating range of 600 to 1400.

V. Likely Conclusion

          Based on all of the information that I have presented, I can conclude that IQ and chess rating and ability have a direct relationship. The most accurate relationship can be found when:

          1) Maximum or near-maximum IQ is found or calculated

          2) Maximum or near-maximum chess rating data has been collected and proven as a pattern

          3) Subjects consistently practice on improving IQ and chess abilities

          4) Maximum IQ ranges between 60 and 140

          5) Maximum chess rating ranges between 600 and 1400

          Since extremely high or extremely low IQ’s and chess ratings can be inaccurate (although researchers are consistently working on improving the accuracy of tests to indicate these extremes), a more detailed and accurate correlation between chess rating / ability and IQ may not be established readily. But since a majority of players would, at some point, fall in the ranges of IQ and chess rating aforementioned, the general correlation between chess and IQ should be found.

          Based on this general information, progress involving chess and general intelligence (or at least some factors of intelligence) can be further determined. This could lead to developments to help children increase their abilities to deal with problems outside of chess and IQ tests. I hope that my hypothesis and analysis has solved some long-argued problems and can lead to these developments of intelligence.

Thank you very much for reading my article. Please leave comments; if you argue against my points, please use some references. If you wish to provide extra and new information supporting my points, also provide some evidence.

References Cited:

http://www.assess.nelson.com/pdf/sb5-asb1.pdf (In part II-B)

http://www.notebook.kolchenko.com/intelligence/blind-chess-and-working-memory (In part III-A)

http://www.auschess.org.au/articles/chessmind.htm (In part IV-A)

http://www.jlevitt.dircon.co.uk/iq.htm (In part IV-C)

junk article

Thanks for taking the time to post your article, which I found interesting (though perhaps a little verbose).

One correction is "Weschler" should be spelt "Wechsler" (same mistake people generally make with "Fuschia/Fuchsia").

kenneth67, thanks for the point - I'll try to find it and correct it.

Edit: Corrected the error.

panandh, please explain why you call it a junk article.

Because he couldn't understand the first paragraph, obviously. Its long, but I enjoyed it. Any chess player that wants to improve needs to understand this to some extent.

HrodHerick, the compliment is appreciated.

However, I'm still looking for some constructive commentary that could either further support or disprove my conclusions here.

I'm sure there are members on this website who have attempted to figure this issue out as well.

I'm going to give this a small bump because this little troll (KILLCHRIST) is having too much fun being an amateur troll. The best way to get rid of him is to ignore him.

I digress, however. Back to the topic if anybody has the will power to read this long article. :)

Interesting article/paper. I think that what is generally referred to as intelligence is a group of mental abilities measured by a test or success in a particular endeavor. A person with a knack for design and an aptitude for mathematics might become a successful bridge builder, for example.

I think its the same with chess, and your paper identifies some if not all of those abilities. I also agree that getting chidren into chess progarams will help their academic progress, especially math. I think that it could also help keep your mind sharp as you age. Well done!

With regards to the elderly, it has been established that the intelligence level becomes "crystallized" due to insufficient/lack of stimuli, and deterioration of neural pathways/loss of brain cells (point #2 and #5 in your chapter   " B. The breakdown of IQ according to its developers", i.e. fluid reasoning and working memory would be affected).  Mentally stimulating games like chess could, I am sure, delay/halt this deterioration. Did you come across further insights on this in your research?

I believe it would be greatly beneficial to students of psychology if IQ tests were more frequently undertaken (say every 10 years) so as to monitor the point/s at which intelligence begins to deteriorate, or improve (taking into account negative life-changing events like losing a partner/ losing one's employment, illness, etc..., and positive changes such as playing sport, joining a chess/bridge club, better diet, etc...) We all get tested in school at about age 16, but when else? I went for a Mensa test when I was 37 years old (am now 43) and would find it interesting to go again when I am, say, 47 - to see if there may be any significant difference (hopefully positive). By defining the most likely age where one's mental abilities will start to decline, it is possible that the decline could be counteracted by a positive change of lifestyle, and playing chess could be one of the simplest ways of doing this.

I think that chess and iq are connected very closely, but iq and rating are not... chess is game that needs  lots of practice, and many of players on chess.com are beginners... So you first losing and your points are gong down and when you get into the game a lil' bit, it take time to reach some good level of points... that is my case, my rating now is about 1100 and i'm winning against opponents above 1300, 1400, 1500 pts. So, your theory is not correct... 

Hmmm...Personally I am now scoring 87 not out; but although short term memory is a bit of a problem I can now see better than ever before the nature of the follies and manifest evils of  of mankind as daily displayed, so experiece needs to be factored in to the equation and it may be that increasing longevity in democratic developed nations may eventually be important beneficial factor  off setting its costs in healthcare etc.

Rybka... IQ.... LOL.

Rybka will surely beat me in chess, but it cannot beat me in kick boxing. (By the way what is the rule for kickboxing, I would like learn and defeat rybka )

I feel if a person has the minimum intelligence of understanding the moves of chess pieces on the board, then any further intelligence will not help him much in playing the game better. After having that minimum intelligence, what he needs is talent for the game - the kind of talent where you give one quick glance to the board and immediately see the right move; more as a reflex than deep analysis.

Thank you for the constructive commentary, in particular from kenneth67. I haven't done research on this topic for roughly 16 months, so I'm not going to bring you any new conclusive information for a while. :)

However, it seems that some people have misunderstood the points I had already established. Closer reading would clarify the objections you've made. Intelligence, as already mentioned, consists of different factors, and talent would definitely be included within one of those factors. The ability to think deeper into chess moves (beyond the initial understanding of pieces, etc.) does require a higher form of thinking which is classified by certain forms in the IQ test (along with other parts of intelligence). As for the argument about Rybka - it must be considered that it is programmed by humans (most often very strong chess players), and that it uses a different system of figuring out the best moves than do humans, thus the noticable difference in gameplay. And, as mentioned before, chess-playing ability does stem from a certain amount of intelligence, a small part of which is measured by IQ.

I'm still looking for one strong form of proof disproving the conclusions I've come to. I would very happy if somebody would be able to find such proof and post it here.

thanks for posting the article.

NatefJay, thanks for the constructive comments. I should confirm the fact that I did NOT state that chess is the ONLY way to improve the IQ factor of a person's overall intelligence. However, since this article is about the relationship between chess and IQ, I thought it would be appropriate to show that practicing chess (which uses factors of IQ) should be a strong factor in improving it, even if by only a small amount. :)

Any activity that stimulates the brain to work in fields associated with IQ will likely improve IQ: it's equivalent to practicing memorization so that your memory improves. And there are a number of methods to do all of those. Chances are, it will happen. :)

I see many people of below-average IQ and above-average IQ playing chess. I wouldn't know if they have reached their maximum potential, but it's usually fairly evident when a person could be considered medically 'mentally retarded' or 'genius'. From what I've seen, my understanding of the relevance seems to be fairly accurate, and I still have not seen any extreme exceptions (although I'd be very intrigued if anybody would point one or more out).

Don't argue with this guy. He used a lot of big words so he must know what he's talking about.