Thursday, April 19, 2007

Little Bright Lies

Staffan A. Svensson headhshot by Staffan A. Svensson

An article I read recently claimed that "research on how junior high school students' beliefs about intelligence affect their math grades found that those who believed that intelligence can be developed performed better than those who believed intelligence is fixed"[1]. This result provides an example of a choice that many people face every day: do we teach what is beneficial or what is more true if these two contradict each other?

I'm not talking about taking it to extremes as in telling someone they are dying to teach them to value life, or the tyrannical ruler kind of "people don't know what's best for them"-attitude. I'm talking about examples like the one above, or telling a teacher that the students are above average in intelligence (which also tend to raise scores[2]), or exaggerate just how bad smoking and eating junk food really is. Or telling someone you can accomplish anything if you put your mind to it.

Sure, sometimes it just comes down to what is called little white lies, like telling someone they look good to make them happy. But I believe my opening example illustrates something more than that; it is choosing to deceive for that persons own good. And there's the rub, because somewhere in there hides a more or less personal, and more or less distinct, line between what is acceptable and what is not, between what is justifiable and what is not.

Personally, I would prefer to be fooled with regard to the world being fundamentally a nice place even if it actually is not, and I intend on making that happen.

References:

[1] online article "Students who believe intelligence can be developed perform better". http://www.eurekalert.org/pub_releases/2007-02/sfri-swb013107.php

[2] Rosenthal, R. (1998) Covert Communication in Classrooms, Clinics, and Courtrooms. Eye on Psi Chi. Vol. 3, No. 1, pp. 18-22. http://www.psichi.org/pubs/articles/article_121.asp

Read More...

Monday, January 22, 2007

Mnemonics and Blindfold Chess

by Staffan A. Svensson

Introduction

Mnemonics can be defined as "a technique of improving the efficiency of the memory"1. It sometimes means "a system to develop or improve the memory"2, implying a specific set of routines to achieve this improvement, but the first and wider definition is the one used here.

The goal of this article is to briefly describe what mnemonic techniques are and a few ideas regarding how they might be used when playing chess blindfolded. The research has consisted of articles and e-mail exchanges with the very competent blindfold player Hindemburg Melao Jr.

Footnotes are used exclusively to indicate sources or cross-references; a chess player is referred to as "he". Thanks to Andreas Gunnarsson, Eliot Hearst and Hindemburg Melao Jr.

What is mnemonics?

The memory processes

Memory if often divided into four major processes, and I will use these categories to help explain my view of what mnemonics are.

Attention and selection — what you notice; you choose (consciously and/or unconsciously) what to focus on.

Encoding — what you have chosen to focus on is changed, encoded, into the things to be remembered.

Storage — how you hold on to the information; some memories fade faster, others slower.

Retrieval — recalling what you have previously stored.

Some mnemonic guidelines you know from your common sense. For example: with regards to attention you need to focus. When it comes to encoding you need to keep it as simple as possible. What is the most effective way of storing (visually, audially, kinesthetically, etc.) differs somewhat between persons. The variant you use spontaneously is probably a good clue. Retrieval depends not only on how well the memory is stored, but also on what you have to do to remind yourself of it (for example, sometimes you forget why you walked into the kitchen, but you know that you can do the walk over again and it will probably come back to you).

Mnemonics builds on this and lets you be more efficient in how you use your memory. It is basically just a continuation of the common sense, taken to a level many people do not bother with because they normally do not need it in their everyday life. It is all built on principles natural to us, simply because these are the ones we do best.

The role of mnemonics in the different memory processes

To improve the attention and selection process you can practice concentration, both intensity and stamina (some form of meditation is often used to achieve this3). In this process we also include everything that has to do with creating an environment suitable for concentration. This includes external factors such as avoiding any disturbances and being given new information in a way that is clear and that you are comfortable with. It also includes internal factors such as being in good health, rested, relaxed and without any feeling of (negative) stress or pressure. In this process is also something seldom mentioned: the question of what you are supposed to pay attention to. The reason this is so rarely included in any mnemonic guides is of course that it is very subject specific.

Just to be clear: mnemonics means the deliberate use of ways to improve these factors. We can all concentrate more or less and what we do without having to actually think about it is not included in mnemonics. It is true that continuous use of mnemonic techniques will incorporate these into your normal thinking. However, when that happens, they are no longer mnemonics.

The encoding process is where you find all the famous mnemonic tricks that make up so much of the self-help literature on this subject4. So how can we make this process as effective as possible? We start by making the information we must remember as simple and logical as possible.

By organizing the information we lessen the amount to be remembered. We do this by distilling from our sources what we actually have to remember, we look for patterns and we decide how much of what we have left that actually has to be memorized. Not all of the original information needs to be encoded, you just need enough to remind you. Once you have found the memorized cues, you can often take it from there and remember the rest. So how do we encode what we have left?

Because we are different, the methods most effective to us differ as well. But there are still some general principles that seem to apply to practically everyone. One of those is that it is easier to imagine something concrete. A concept or anything else abstract is transferred into something concrete, which is remembered (concrete means you are able to sense it; see it, hear it, smell it...). In this and the other encoding situations imagination plays a big part.

Another basic concept is that it is always easier to remember something that has a clear connection to something you already know. Association with something familiar gives you a specific place to put the new stuff, a place where it is easy to find later. In order to make the associations as rich and effectual as possible, it helps to use all senses. Not just see an image but think about what it sounds like or what it smells like, or anything you might imagine. It is also good to attach some form of emotion or mood (if that doesn't come by itself); an emotional event is easier to remember than one you don't really care about.

A note about automatic encoding (also called "chunking"): In practically all aspects of life we use what is called implicit knowledge to automate tasks we perform regularly. For example, you do not have to think about how to walk, how to talk or how to read. It comes automatically. As mentioned above, this kind of simplification of input is not included in the term mnemonics.

Storage itself is not subject to mnemonic techniques, but the result of the other processes.

The retrieval is, because it is the decoding process, inevitably linked to the encoding process. Whatever you have associated with the memorized information is your key, so that is what you use in finding it again. If there is still something you cannot remember, the only thing you can do is search for it.

If you have something "on the tip of your tongue", that is if you know you have the information but cannot access it, you can in a limited way still look for cues. If it is a specific word, like a name, you can look for it by trying to start the word with the letters of the alphabet, one by one. Hopefully you will be reminded while trying the correct letter.

This sort of retrieval help, which is really just a form of systematic search, is only the last resort and not very effective. When developing the mnemonic techniques, all the work goes into the attention and selection and encoding process (in other words the input processes).

How can mnemonics be utilized in blindfold chess?

Remembering one board

Some parts of mnemonics are always present, no matter if you use a specific system or not. They are the ones in the attention/selection process. You have to be focused and you have to pay attention to only the game. All the factors regarding focus and concentration mentioned above apply.

In the encoding process the simplifying of information is always there, mostly spontaneously. This is what has been called "chunking" of information. This is necessary if you are to handle all the information needed to play a chess game, but since it is automatic, it is not included in mnemonics. (It is possible to focus on this part of the process while playing, but that counteracts itself since the goal is to have less to focus on, not more.)

Chess players don't remember a position piece by piece, as a non-chess player would be forced to. They see relationships between squares, pieces, pawn structures, open files, and so on. The better the player, the more efficient the chunking of the information of what the board looks like, and the more elaborate the associations of squares, pieces and piece configurations. For someone who doesn't play much chess, playing blindfolded sounds like an enormous mnemonic effort, but it is much simpler for someone who has the tools for it. Being a good player means having an efficient set of tools.

When it comes down to it, remembering the game, as in remembering all the moves in their correct order, is not the same thing as being able to "take in" more or less the entire board at once. This kind of comprehension is required because it is the only thing you have to go on to calculate your next move; just remembering what piece moved where is not enough. If you have enough knowledge and skill at playing the game with a board that you can plan moves without it, then remembering what you did is not a problem. It is then not much information to remember. This means that playing blindfolded is not really a question of having a good memory, is it about being able to comprehend the position enough to be able to plan your next move. In a way all players use this ability more or less even when they have a board in front of them; while planning ahead, they envision pieces moving and watch for what kind of position the moves lead to.

In blindfold playing, there is a skill level below which the information gets too complicated for the brain to process. Master blindfold player Reuben Fine (1914-93) has written he believed that knight odds level is required to play one game blindfolded, while master level is necessary to play more than one game 5. As people differ in there working memory capacity, the required level probably correlates with both size and configuration of that capacity, as well as with ability to concentrate. Examining this and getting more information about at what level of chess skill blindfold playing is possible would make interesting research but I have found no more information than the above cited article.

There have been suggestions how to remember chess games even if you are not a competent player. Dominic O'Brian has suggested using a variant of the Journey method6. He gives the different pieces personalities and the Knight so becomes Sir Lancelot of the round table, the Queen is Elisabeth II etc. Then algebraic notation itself is made able to visualize. Each square is turned into initials by changing the number into a letter, c3 becomes CC (represented in his example by Charlie Chaplin) and f6 FS (Frank Sinatra). The different images are then associated with each other and stationed along the mnemonic itinerary. This way of memorizing results only in recollection of moves in their correct order. It does not relay any relationships between pieces, which means that no matter how many games you remember, it will not improve your playing.

We come to the conclusion that some general parts of mnemonics, more specifically the ones you can benefit from in any type of situation, help if you want to play blindfolded. More specific tools, such as the systems found in many books, do not.

Remembering more than one board

Let us assume a person can play one blindfold game of chess. How can he go about if he wants to play more than one game simultaneously?

Playing more than one board could be seen as doing the exact same thing as with a single board, but with the amount of information to be remembered multiplied with the number of boards played. It could also be seen as two different and separate activities, playing (comprehending) one board and remembering the others.

In the first alternative the skill level required must be significantly higher than when playing only one board. In the second alternative, the player uses the same way of playing the single board as he has done when playing only one game. The factor added is to put the other board or boards aside and recall it for the next move; that is, simply remembering something enough to be able to recall it later. I say "simply" because this process does not require this board to be available to plan moves or strategies, it just has to be stored. The issue of being able to use the board for planning moves is still only needed for one board at a time. The other boards are stored, ready to be picked up again and played, and this storing is the area of the type of mnemonics featured in popular mnemonic systems (a number of which are listed in the Appendix). In reality, however, you will not find anyone using exclusively one of these two alternatives. They represent only the extremes of a spectrum.

The more you use a mnemonic technique, the more automatic it gets. You simply get better at doing it as you chunk the steps involved better and better and eventually it gets fully automated. A good example of an automated process is the way you read. You probably do not have to think about what the letters or even the words mean, as you were once forced to. But it is not as simple as that, and rechunking happens many times during the learning process. What is to be chunked changes as comprehension of it improves, and comprehension improves when the new chunks are organized. I will not try to explain these developments, but I will look at where mnemonics can be applied.

When you recall something you start with one detail and that detail reminds you of another, and then another, and soon you have more or less the complete memory. And even if some detail is missing you can probably find more than one path of association to remind you of it once the others are in place. Accepting this model of our memory we get two places where mnemonics can do their thing: helping us find the first clue, the "key," and helping associate the pieces of information so that the key will lead to all the rest.

The associations between the information involved in a chess position are always more or less spontaneous, since the moves follow a specific plan and this binds them together. The best way to help a player to improve is probably just to remind him of the basic guidelines of mnemonic associations and let him do the specifics (I say this because personally adapted mnemonics are always the best, and I am careful not to try to improve on the spontaneous by applying a general model). The guidelines, described briefly in the first part of this article, are: use your imagination, look for patterns, use all your senses, associate with something familiar, use concrete images. All these rules apply also when remembering the key bit of information, but as this is a more straightforward memorizing, mnemonics can here be given a larger and more elaborate role. For this, one can use any of the systems described in a number of books on mnemonics and memory.

Making the boards distinct

The most common problem described in simultaneous playing is that of mixing up the boards. If you only play two boards then this will not be a problem since you will be working on one of them at any given point in time. But how do you keep the boards separated when playing ten boards or more? By making them different.

A common technique is using different openings to separate boards. When playing twelve games, the blindfold player use one opening on four of the games, another one the next four and play black on the last four (or some similar system)7. Some people think that the memory used by blindfold players is built up by a memory bank of "normal" positions. This, they conclude, would make the game easier to remember if it did not include any unexpected strategy or odd moves. If they try these in order to make the player forget the game easier, the effect becomes the opposite: that game is immediately singled out from the others and thus easier to remember8.

In this way it is a matter of time before games become distinct enough from each other that there is no chance of mixing them up. The problem is the player must at all times have the distinctiveness of the games clear enough, even if only one or two pieces separate them. This kind of situation not only lends itself to mnemonic systems but to a particular type of system called "loci". Loci means place and the system consists basically of positioning that to be remembered in different surroundings already familiar to you. In your mind you already know your way around many locations, separate not only in space and time but also with regards to the feelings you associate with them; these locations can be used when profiling boards. The way the locis are used must depend on how the game is already comprehended and remembered by each player, but here are just a few suggestions:

If the game resembles a normal game in the way it is envisioned, why not play one game at your kitchen table, one at your favorite chess club, one in the park. Use familiar places and odd places, even imaginary places. If your view of the game is more of a story unfolding then let it play out before you on different stages. There are no rules when it comes to this and weirder is often better. Another way to approach it is to imagine familiar historical personalities as opponents, why not Napoleon or Sun Tzu (author of "The Art of War").

I am unfortunately not a good enough chess player to test these ideas in practise so their utility is so far not much more than a guess. However, some prominent blindfold players have been able to perform tricks worthy of any mnemonic expert9, so there are very likely already quite a few productive ways to use these systems. My humble suggestions above will of course not work for everyone, but my hope is that it will help someone.

Appendix - Common mnemonic systems

Below are a list of the most popular mnemonic systems. It is easy to see that many of these are variations on the same themes and they often overlap. There are many good books describing these and the way they can be used. The names used are the most common according to the many sources I have checked when compiling this list.

Link — you link together each thing to be remembered in an often-absurd story.

Substitute word or phrase (also called Keyword) - if the thing to be remembered is not easily visualized you substitute it for something else.

Number/rhyme (Pegword) — use something that rhymes with the number, for example 1 = bun, 2 = shoe. This is then used to get a numbered list by associating a shoe with something you know is item number two on a list.

Number/shape — instead of using something that rhymes you use something that looks similar to the number. 1 becomes a stick or a candle, 2 becomes a swan etc.

Major (Figure Alphabet, Phonetic Alphabet) — each number is represented with one or more consonant sounds. 1 = d or t, 2 = n, 3 = m, 4 = r. The number 43214 becomes for example "reminder" (r-m-n-d-r).

Alphabet — some image is connected to each letter instead of number, usually representing something that starts with that letter. A = ape, B = Bee, C = Sea etc.

Loci — the images to be remembered is placed in a certain location (= loci). For this you use a real building or route you are familiar with, either real or imagined. When memorizing or recalling you imagine walking the same route.

Journey — an extended loci system that often includes travelling between known locis. Sometimes denotes the same system as loci.

Roman room — once again thing placed in a location, usually a room, this time not in any specific order.


Footnotes

1 Webster's Third New International Dictionary — Unabridged

2 The American Heritage Dictionary, fourth edition

3 O'Brian, D. 2001, Lär dig minnas (Swedish translation of Learn to Remember).

4 The most common mnemonic systems of this kind are listed in the Appendix.

5 Fine, R. 1965. The psychology of blindfold chess. An introspective account. Acta Psychologica. 24: 352-370.

6 O'Brian, D. 2001.

7 Binet, A. 1966. "Mnemonic virtuosity: a study of chess players." Genetic Psychology Monographs. 74: 127-162.

Cleveland, A. A. (1907)." The psychology of chess." American Journal of Psychology. 18, 269-308. Fine, R. 1965.

Personal e-mail exchange with Hindemburg Melao Jr.

8 Fine, R. 1965.

9 Cleveland, A. A. (1907).

Read More...

Wednesday, November 22, 2006

Another Puzzle

Staffan A. Svensson headhshot by Staffan A. Svensson

The pieces of this puzzle should be combined in a way that will be the obvious answer when found.

Look below for the solution.

Solution

puzzle solution

Read More...

Thursday, October 26, 2006

Puzzle

Staffan A. Svensson headhshot by Staffan A. Svensson

The pieces of this puzzle should be combined in a way that will be the obvious answer when found.

the puzzle

Look below for the solution.

Solution

puzzle solution

Read More...

Wednesday, October 11, 2006

Strange Ideas

by Staffan A. Svensson

"I think I can safely say that no one understands quantum mechanics…. Do not keep asking yourself, if you can possibly avoid it, "But how can it be like that?"… Nobody knows how it can be like that" - Richard Feynman

We humans have certain intuitive ways of seeing the world. We can be said to have a kind of intuitive physics, seeing the world as consisting of objects and being able to keep track of how an object falls and bounces (or breaks). We could also talk about a sort of intuitive engineering used to make tools and to understand how they work. There are others, such as a spatial sense to navigate the world with, a number sense, and a basic sense of probability. All of these are very useable to us and of course based in evolution. Unfortunately this intuitive thinking is better adapted to small groups of illiterate people who live off the land and depend on what they can carry than to todays society.

As much as these abilities have helped us survive and to overcome real world problems in the past, they also per definition limit our thinking. When it comes to understanding, these modern times practically necessitate dealing with objects on a daily basis whose inner working is a complete mystery to us. I have no idea how the computer I am writing on really works (and it makes a point of proving that on a regular basis), I do not know how my washing machine or even how my front door lock works, or what makes my wristwatch tick. But somebody does, the principles all these things are based upon are pretty basic, and the point here is that if I wanted to I could still learn it and understand it. Now instead take quantum mechanics, mentioned in the introductory quote. We have every reason to think these theories are true, but at the same time they go against many of our intuitions about space, time and matter. Even the experts now say nobody can understand the present theories. Does that tell us anything about how the possibilities of developing these ideas? Does it mean that we might reach a point where we can't develop our theories further because we don't really understand them?

If so, it is not really a question of complexity, but about incompatibility. We already have computers taking care of the complex calculations, and there seems to be lots of room for improvement on that front. Incompatibility with the way we are pre-wired is something different. You can't brute-force yourself out of it since you have no idea where to go (the expression of having "no idea" actually captures the point quite well). It is inevitable that the built-in restrictions prevent us from seeing a lot of things, and nothing says evolution has made us see the world around us more and more accurately.

But thinking about this subject, I can not help being reminded of the sometimes famous and historically reoccurring quotes from people saying we will soon have nothing more to invent, that ideas are running out. Fortunately, these estimates have always proven hilariously wrong, and I see no reason why this instance would be any different.


[The strange idea of writing this article came from reading chapter 13 ("Out of Our Depths") of Steven Pinker's book "The Blank Slate".]

Read More...

Tuesday, October 10, 2006

Absolute Pitch

by Staffan A. Svensson

Have you ever wondered how some people are able to sing a specific note on command, simply tell you "that note is a G flat" or "the key of the song is C"? These people are using something called absolute pitch, and in this article I try to briefly describe what it is, where it comes from and how you probably have some of it yourself, without even knowing it.

Absolute pitch (sometimes called "perfect pitch") contrasts to Relative pitch, and both are used to describe how a person hears notes, or more specifically pitches. These terms are usually used in music, but are equally valid whether you are talking about music, engine sounds or birdcalls.

When someone uses relative pitch, they compare one pitch to another pitch, so to be able to use relative pitch one needs a minimum of two notes (or some other sound with a pitch). When someone hears that a melody first goes up and then down or that the pitch goes higher when an engine accelerates, they are using relative pitch. This type of listening is taught in music schools and is the basis for all types of intervals, scales and chords. Almost all people have some level of relative pitch, and even those who are tone deaf (approx. 4%[1]) are not completely without this ability, even if it is markedly reduced.

When someone uses absolute pitch, they are listening to what frequency a tone or a group of notes have. You only need a minimum of one audible pitch to use it; it is not about comparing it to anything. This has for a long time been seen as a sort of almost magical innate ability to name a pitch or to sing a specific note on command. Some have seen it as a curiosity with little practical significance, while for others it has been the holy grail of music excellence. Lack of knowledge has made the whole concept a well-nourished growing ground for rumours and guesses.

Relative pitch has never been a problem for people to understand. Almost all people hear the difference between a high note and a low note if you play one after the other, and some people do this better than others. No mystery there. But what of this weird absolute pitch? What so few realized is that some people hear frequency differences better than others. In the same way people see the differences in frequency when we see colours, we also hear them (we can only see about one "octave" but a person with good hearing can hear 10 octaves). But everyone that has potential to see colour seems to naturally develop the ability to see differences in colours, while few people seem to hear the frequency differences. Why? Part of the answer, as with many other hard questions, comes with defining what we really mean by the question. If hearing colours means a person can identify tones played on any instrument and name them correctly each time, then very few will qualify. If hearing colours instead means someone can hear that a melody played in one key sounds or feels different in some way, or has a different "flavour" to it, compared to the exact same melody played in another key, suddenly this group includes most people. Such an experience indicates the same kind of ability, only not refined enough to be able to differentiate between and thus identify semitone intervals as standardised in western music.

The American Heritage Dictionary defines absolute pitch as the "ability to identify any pitch heard or produce any pitch referred to by name"[2]. Other definitions include "capacity to identify any tone upon hearing it sounded alone or to sing any specified tone"[3] and "that sense which some people possess of the actual pitch of any note heard"[4]. There exists no word for weaker forms or for this ability as a whole, and this may partly explain why so few people realize the simplicity of the phenomenon.

So why doesn't every person possess excellent absolute pitch? It does not seem as important to us as seeing colours, so even though the ability to learn is innate, practise is required to get it to the level where one can use it in music. Most people do not have that need, so the ability never gets very cultivated. Many have no idea that the possibility exists, and if they do not happen to discover it for themselves they will never even try to develop it. This accidental discovery is probably what causes some children to develop it spontaneously.

One theory suggests absolute pitch evolved to "subserve speech, and that it may be readily acquired by the association of pitches with meaningful words very early in life, during the critical period in which infants acquire the main features of their native language"[5]. Tonal languages use this particular feature of language more than many western languages, including English, so do people with a tonal first language have a better sense of absolute pitch? It turns out that they do, and that it carries over to music. Deutsch et al. tested Chinese and US music students ages 17-34 for absolute pitch (defined as at least 85% correct answers identifying individual piano notes) gave results like these: "For those who had begun musical training between ages 8 and 9, roughly 42% of the Chinese students met the criterion whereas none of the US nontone language speakers did so."[6] It has been known a long time that most of the people with a good absolute pitch started their music training when very young, and this certainly seems to indicate if not a critical then at least a sensitive period for this type of development.

At this point you might ask yourself if you have some noticeable level of absolute pitch, so why not try it for yourself? You may not have the opportunity to listen to the same musical piece in different keys in order to compare them unless you or someone you know can play them, but if you have an instrument it is easy to pick it up and listen to individual notes. Play long notes and take your time. I would not be surprised if you thought the tones do not actually sound exactly the same. Some notes may sound a little "softer", others a little more "twangy". Putting it into words like this is like describing colour to someone totally colour-blind, and some people prefer instead to call the notes by colours (and no, those who do seldom agree which note is which colour). Also, some of the subtle differences you might hear could be because of the instrument (the different strings on the guitar will not sound the same), but these are usually easy to spot. The important thing is to take your time. Good luck!

References

[1] http://www.annalsnyas.org/cgi/content/full/1060/1/311, Sloboda, J.A., K.J. Wise & I. Peretz (2005). Quantifying tone deafness in the general population. Ann. N. Y. Acad. Sci. 1060: 255-261.

[2] The American Heritage Dictionary of the English Language, Fourth Edition ©2000, updated 2003, Houghton Mifflin Company

[3] http://www.reference.com/browse/columbia/absolpit, quoting The Columbia Electronic Encyclopedia Copyright © 2004, Columbia University Press

[4] http://www.highbeam.com, quoting The Concise Oxford Dictionary of Music, 1996

[5] http://psy.ucsd.edu/~ddeutsch/psychology/deutsch_research9.html, from the homepage of Diana Deutsch, Ph.D.

[6] http://www.aip.org/148th/deutsch.html, Deutsch et al. (2004). Perfect Pitch in Tonal Language Speakers Carries Over to Music, popular version of paper 3pMUb3, 148th ASA meeting, San Diego, CA


If you are interested in developing absolute pitch, this is the most pedagogical and well worked-out source I have come by so far:

Burge, D.L (1999) The Perfect Pitch Ear-training Supercourse. Information can be found at http://www.perfectpitch.com/

Read More...