Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
It's funny.  Laugh. News

The Memory Masters 282

Vaystrem writes "Wired's Article 'The Masters of Memory' details the outcome of the recent U.S. Memory Championship ,where 'three dozen people who had, in just five minutes, memorized the positions of 52 cards in a shuffled deck and were now happily organizing cards in a new deck into the same order as the pack they had memorized.'" The article includes details of "the mind numbing upcoming world championship. Could you in a half hour 'memorize a random string of thousands of 1s and 0s'?" I'm still working on the mnemonic alphabet.
This discussion has been archived. No new comments can be posted.

The Memory Masters

Comments Filter:
  • computers (Score:5, Funny)

    by panxerox ( 575545 ) * on Saturday March 06, 2004 @04:50PM (#8487146)
    I'm a geek, memory is what I use computers for so I don't have to. (besides HD mem storage dosent frag out after a hard weekend and a keg of beer)
  • by Smitedogg ( 527493 ) on Saturday March 06, 2004 @04:51PM (#8487153) Homepage
    Crap, what was I going to post about?
  • by Space cowboy ( 13680 ) * on Saturday March 06, 2004 @04:52PM (#8487163) Journal
    (the quote is limited due to the size of the heading, but 10 is right out!)

    The brain seems to actually have the sort of grasp of numbers that we sometimes ascribe to "Neanderthals" ... It really does seem to go something like 1,2,3,4,5,6,7,lots. We seem to have a distinction of the innate "three-ness" of a scene, for example, and don't need to count to know that the quantity of X is three.

    Different people vary with the maximum innate value they just grok, with most people coming in about 5 or 6, rarely do you get 7, and vanishingly rarely do you get 8.

    What has this to do with memory, you cry! Well, in the same fashion, we can innately recall small numbers of things, without doing an exhaustive search. This is useful for PIN numbers :-) The thing is that we can do it recursively, with a bit of effort, so you remember group A is (21,63,37,78,39) and group B is (25,544,62,150,311). It's easier to recall both sets if you first subdivide into the largest quantum you can most-easily recall, and remember the sets individually. Normally you can do this for the number of sets in your personal quantum, so if you can easily remember 5 numbers in a set, this helps you remember 25. It's not "free" of effort, but it's a lot easier than remembering 25 numbers straight off..

    Hack the system! exploit the underlying nature of your brain!

    Simon
    • by LostCluster ( 625375 ) * on Saturday March 06, 2004 @05:07PM (#8487259)
      Where can I download the patch for the faulty entry in my long division tables?
    • One of my psychology books told the story of a world champion mnemonics person who did hack his grasp of numbers and alphabets. He quite sadly recounted the story of how he cannot now read a book without every letter bringing up some string that he has remembered in the past.

      After I read that I desperatly avoided mnemonics.

      • The story you are referring to is the true account of one the most famous subjects in psychology--"S" studied by the Russian neurologist AR Luria. He authored a book called "The Mind of A Mnemonist: A Litte Book About a Vast Memory." The man could not forget anything and was tortured his whole life by it. Highly recommended reading.
    • The number is 7+/-2 (Score:5, Interesting)

      by NoData ( 9132 ) <`moc.oohay' `ta' `_ataDoN_'> on Saturday March 06, 2004 @06:12PM (#8487692)
      Perhaps the most famous, certainly one of the most cited, papers in cognitive psychology is George Miller's 1956 paper "The magic number seven plus or minus two: some limits on our capacity for processing information." The 7+/-2 rule is one of the few, true "laws" in psychology. It describes the number of items that can be held online in working memory by the average individual. I won't even begin to touch here the myriad theories that proposes mechanisms for this limited capacity.

      The technique you talk about regarding the grouping of multiple memoranda into a single unit is called "chunking" and was studied by another great in psychology, the late Herb Simon of CMU. He and Bill Chase found that chunking was basically what set chess masters apart from novices. They saw entire board configurations at once, rather than the relation of individual pieces.

      The ability to appreciate the numerosity of multiple items without counting is called subitizing. I know less about this, but the average person can subitize up to about five items.

      Anway, just wanted to give credit where it's due for what has become pop psychology fodder.
      • by DJStealth ( 103231 ) on Saturday March 06, 2004 @10:31PM (#8488960)
        A good way of chunking 'a string of thousands of 1s and 0s' would be simply to convert it to Hexidecimal. A string of 1024 bits can easily be compacted into 64 hexadecimal characters. Maybe if you're lucky you'll have it spelling things like DADA D00D F00D 1234

        Chances are it won't be that easy, but you get the idea..

        Similarly, most people thing their fingers are only capable of counting to 10. But how many of us think to use our fingers to count in binary to reach 1024 (2^10) possible values?
        • by jtdubs ( 61885 )
          Sorry dude, but 1024 bits is 256 hex digits. Each hex digit represents one nibble, or one half-byte which is 4-bits. And, 1024 >> 2 is clearly 256.

          Now, I notice you are listing them in groups of 4, which makes 64 groups of four. But, those groups aren't hexadecimal characters, they are hexadecimal words of 2-bytes in length.

          Justin Dubs
    • Fascinating.

      How does that relate to visually counting items? I'm not a savant, but where I work I've had/developed an ability to count large numbers of items by what I could call the "two sets of five" method; if I'm doing inventory I can count items, without actually sorting them, by the ten - I 'see' two sets of five, the next two sets of five, etc - brain processes 10 10 10 10 5 1 = 46 - enter it in the Telzon and next batch (yes, I do inventory control, but it pays well :) especially if you have met
    • The number is not seven, it is seven plus or minus two. The originator is George Miller. The theory is information processing. The reference is: Miller, G.A. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63, 81-97.
      [Available at http://www.well.com/user/smalin/miller.html]

      The concept of putting more things together in groups to remember more than seven plus or minus two is called "chunking". The telcos paid close attention
  • Hmm (Score:5, Funny)

    by mnemonic_ ( 164550 ) <jamec@u m i ch.edu> on Saturday March 06, 2004 @04:53PM (#8487168) Homepage Journal
    I feel obligated to reply to this story.
  • by mikeophile ( 647318 ) on Saturday March 06, 2004 @04:56PM (#8487190)
    The Vegas Casino Consortium. All winners will receive lifetime bans in every casino in the world.
    • Re:Sponsored by... (Score:5, Interesting)

      by LostCluster ( 625375 ) * on Saturday March 06, 2004 @05:05PM (#8487244)
      The Vegas Casino Consortium. All winners will receive lifetime bans in every casino in the world.

      Only at blackjack tables. That's the only common casino game where memory of what's happened before matters.

      Some roulette tables actually have displays that show what has happened on previous spins, because any patern you might detect in that data only gives you a false confidence that might motivate you to play, in reality that information is totally useless in helping you predict what will happen on the next spin.
      • Re:Sponsored by... (Score:5, Interesting)

        by pongo000 ( 97357 ) on Saturday March 06, 2004 @06:06PM (#8487653)
        Some casinos welcome "card counters." I have a friend who lives in Santa Fe and is an extemely good card-counter -- he can memorize cards dealt from a 3-deck shoe. Having a player sitting at a blackjack table raking in chips is a draw...no one wants to play at a "losing" table. So my buddy, he makes money on the side playing blackjack, and even thought the casinos lose on him, they make money on the ones that come over to play at the "hot" table.

        The idea that card counters are not welcome at blackjack tables is a myth. Instead of fighting the problem, they now have figured out a way to make money on it.

        As an aside, this guy is an air traffic controller (I used to be one as well). Most air traffic controllers develop an incredible short-term memory, being able to memorize 3-D positions of several aircraft at once in conjunction with an in-memory 3-D representation of the surrounding airspace, available for immediate recall. All of this takes place while listening to a steady stream of aircraft identify themselves with 4- or 5-character callsigns, which are queued up for responses in the order they were received, while also monitoring landlines to various other air traffic control facilities. Not to mention being able to monitor the D-side working next to you talking about his hot night out, as well as the supervisor ranting over your shoulder about bullshit you could care less about.

        The amount of information retained in short-term memory for a moderate to heavy session of air traffic easily exceeds the 104 discrete pieces of static information memorized from a deck of cards.

        Short-term memory only works if it's exercised on a continuous basis. I've been out of that field for several years, and I'm lucky enough to remember a single telephone number at a time.
        • Re:Sponsored by... (Score:5, Informative)

          by geekoid ( 135745 ) <{moc.oohay} {ta} {dnaltropnidad}> on Saturday March 06, 2004 @06:40PM (#8487830) Homepage Journal
          As someone who has done Casino Security(long,long ago) I can say most casinos don't want card counter.

          1 card counter is a draw, 100 card counters is a loss. which is what you get when it becomes known you don;t mind counters. Counter often work in teams.
        • Re:Sponsored by... (Score:2, Interesting)

          by Anonymous Coward

          Sorry, but your friend is either lying to you or not a very good card counter. His concept of a "losing" table is riduculous.

          Since the odds are only about 3% in the house's favor with basic blackjack rules, there will be "winners" at almost every table, just not as many winners as losers over time. In addition, successful card counters try to play at tables with fewer players so that they have more hands to maximize their return when the count is in their favor.

          Good card counters can only tip an avera

      • Re:Sponsored by... (Score:3, Interesting)

        by shfted! ( 600189 )
        Not necessarily with roulette. The roulette table can have a slighty variation which can show up statistically. However, you'd need to keep track of thousands of spins to get any meaningful information. It's just like flipping a coin that showed up heads 9 times out of the last 10 -- I'd bet heads the next 100 times.
  • by Professor Cool Linux ( 759581 ) on Saturday March 06, 2004 @04:56PM (#8487193) Homepage
    It's 42...

    (Sorry I couldn't Resist)
  • by anonymous cowfart ( 576665 ) on Saturday March 06, 2004 @04:57PM (#8487200) Journal
    If technology advanced enough that you could download memories from the brain of someone with extremely good memory, would the brain be an illegal recording device? I read once that your brain can recall almost everything. Some of the material merely needs coaxing out (like with hypnosis). Hmmm....
    • The brain would make a very poor recording device for several reasons. Firstly, the quality of the information stored isn't very good, and deteriorates with time. Watching a movie that you downloaded from some guy who saw it earlier instead of actually going to the theatres would be like listening to a 24kbps mp3 that someone "shared" with you instead of buying the orginal CD. Secondly, and more importantly, you can sometimes generate false memories. See this article [washington.edu] and this article [owt.com]. This is why it's s
    • Gotta love Slashdot posts stolen right out of Johnny Mnemonic...
  • by Anonymous Coward
    I'm would say I'm not so good at strings of numbers or names or anything I have to remember in a short period, but I remember the place of everything in my bedroom. We've done some tests where a friend would slightly move one cd case (out of hundreds) and I could pick out what had changed. I can also remember thousands of songs. Not just the lyrics, but I can replay them in my head like I was hearing them on the radio. I guess these are more natural (hunter-gatherer) than the list-based stuff I'm not so go
  • actually (Score:5, Funny)

    by Digitus1337 ( 671442 ) <(moc.liamtoh) (ta) (sutigid_kl)> on Saturday March 06, 2004 @04:59PM (#8487207) Homepage
    When they're not in competition they're memorizing 1's and 0's for me. I keep them in my basement as a backup in case my harddrives crash.
  • by FiberOpPraise ( 607416 ) on Saturday March 06, 2004 @05:00PM (#8487209) Homepage
    Any system admin would love these guys! Now we can safely create default passwords such as: fG2ajf(Ak&f235Afj!^pt3p%A$2 Without fear of the user writing them down!
    • Easy. (Score:5, Interesting)

      by SharpFang ( 651121 ) on Saturday March 06, 2004 @05:10PM (#8487286) Homepage Journal
      Take some poem, lyrics of a song, some text you know by heart.
      Pick all first (last) letters of each word. Include all punctation marks when needed.
      Convert to 31337 H4X0R speech.
      On some specific pattern (i.e. first letter of every verse) add Shift.
      Trivial to make up on the fly.
      • Re:Easy. (Score:3, Interesting)

        Another alternative is to use a random pronounceable word generator for your language of choice. Entropy/character is lower than that of the first character of random free text, but higher than that of frequently memorized text. (How many Americans who use first letters of a phrase come up with "Fsasya,ofbfutcann" or some initial segment thereof? Answer: lots of people, since they're using the first letters of the Gettysburg Address, which they were required to memorize at school.) Problem is,
        first lett
        • ...unless you introduce some your own "proprietary" way of mangling.
          Say, you attach _ at the end of every password of yours, or use , instead of . and you waste every dictionary-based attack, even being VERY close to the right password doesn't matter. Not in dictionary - won't happen. And the number of possible substitutions... too big.
        • Re:Easy. (Score:3, Interesting)

          by haystor ( 102186 )
          I play "music" on my keyboard. I take some piece of piano music that I have in muscle memory and play it to bang out my password. The passwords are quite good and since it's a slight variation of something I already know, I can memorize it quite quickly. Another key benefit is that it is always easy to type.
      • How about: u472bmt?
    • You bastard, how did you guess my password?

      I'm going to have a busy weekend changing everything to a new one now.

      *grumble*
    • I routinely create passwords like that for myself (usually around eighteen characters); the trick is to *never* write them down, and *never* tell anyone. That extends the re-use level of the passwords.

      For those people I admin, I use a bit of l33t speak... and morr!$ isn't such a difficult password for him to remember, while giving at least a little bit more security. (Yes, I know not that much, but it's a start.)
      • Re:Passwords Anyone? (Score:5, Interesting)

        by SharpFang ( 651121 ) on Saturday March 06, 2004 @05:34PM (#8487456) Homepage Journal
        I prefer passwords written down on stuff.
        Typing this from a Toshiba laptop, with sticker "Contains: Nickel-Cadmium, Nickel-Metal Hybride, and/or Lithium Ion battery" on the bottom.
        The root password is
        c:n-c,n-mh
        Yeah, find my IP and get through NAT. :P
        • I used to use this scheme for web sites (e.g. commerce). I would basically take the url (for example amazon.com) and transpose letters on the keyboard in some way. Then one day one of my passwords didn't work. Turns out the company had been aquired by another company (or vice versa maybe?) and so the company name no longer matched my password.

          Now I just use the Keychain (OS X). I can just hit keys randomly and never see the password, then copy to the clipboard and paste into the browser when I'm ready
  • Mnemesis (Score:5, Informative)

    by SharpFang ( 651121 ) on Saturday March 06, 2004 @05:03PM (#8487233) Homepage Journal
    Open source program for training mnemotechnic memory:
    Mnemesis [sourceforge.net]
  • by Anonymous Coward
    I suffer from CRS..
    (cant remember shit).

    -dirtbag
  • by Fiz Ocelot ( 642698 ) <baelzharon&gmail,com> on Saturday March 06, 2004 @05:06PM (#8487249)
    The article gives the impression that memory = intelligence. But I would beg to differ. So what if you can memorize a long binary string. You may not even know it is binary, nor what the string translates to.

    I guess the thinking is, "well they do very well on tests". Sure, that's because they memorized everything. But do they Understand? There's a difference between knowing something, and really understanding what it means. I really think schools should focus more in testing how well a student really understands a subject, perhaps demonstrate the ability to teach it to someone else.

    • by Gyan ( 6853 ) on Saturday March 06, 2004 @05:12PM (#8487301)

      You bring up the usual objection.

      However, you're limiting memory to declarative memory (where's my keys?, who's that girl?, what movie was that dialogue from?). But your skills are themselves (implicit) memories. You learn when young, something like language. You can construct proper grammatical and meaningful sentences later in life, only because your brain remembers what it learnt earlier.
      • by shawnce ( 146129 ) on Saturday March 06, 2004 @05:40PM (#8487493) Homepage
        However just remembering something that you learned in the past isn't in itself a good indication of intelligence.

        The ability to synthesize new knowledge based on the experiences and knowledge you have learned is a much better measure.
      • No one can deny the importance of memorization. But there are so many factors to intelligence. Face and voice recognition requires both memory but pattern recognition. Not only memory, but proper associations with other things you've memorized. Then you have utilization of that data and turn it into useful information and knowledge. You also have using knowledge to solve problems and new obstacles. There is a whole lot to intelligence and very little we understand of how we do what we do.
        • No one can deny the importance of memorization. But there are so many factors to intelligence.

          Nope, intelligence is memory. Your objection limits what is considered memory. You are restricting memory to objects and treating functions("utilization of data") as something else. But the functions themselves, are memories. You need to store the function neurally, in order to execute it later on in life.

          This is a good introductory [amazon.com] book on memory.
          • "Nope, intelligence is memory." You've made a false tautology from a valid description of the brain. Intelligence is memory, just as mind is brain. Biologically, I agree with you but I'll continue to use those words differently, because they have different meanings. You might just as well claim all our experience is memory, because we only percive things after they are mediated by our sensory organs and conveyed to our mind, brain, and memory. The argument is valid in a descriptive sense, but not definitive
    • Intelligence is not just your ability to make good decisions based on the facts you have, you also need a lot of facts available from which you can make good decisions. People with a good memory and a large base of knowledge from which they can draw tend to make better decisions. Of course the definition of intelligence can be argued.

      Many tests just determine whether you can remember something. Other tests ask oddball questions related to the subject matter, trying to get you to think. But that's really ju
    • Yeah, true intelligence isn't exhibited by someone memorizing a really long string of ones and zeros.

      True intelligence is recognizing instantly upon hearing of the competition that if instead of memorizing it as a long string, you convert the binary to a decimal number, it's going to be a heck of a lot easier to remember. You just convert it back to recite it.

      Why would someone memorize 101010 when they can just remember 42 instead? If hex is intuitive to you already, that's even better for long binary num
    • "You may not even know it is binary, nor what the string translates to."

      if you did, that still wouldn't be intellegence.
      being abole to figure it out with no prior information, that would be intellegence.
  • In medical school we certainly benefited from mnemonics for all sorts of things from cranial nerves to biochemical enzymes, but these guys are on a different level. Of course it could be reasoned that they are able to make other associations that seem logical (perhaps) that enable recall much easier.

    Colors, musical notes, mathematical formulas......whatever makes sense.

  • These guys are remarkable, no doubt about that. But the main reason that they are able to have such phenomenal memories is that they can easily come up with quick and easy pheunomics so they can remember things like orders of cards, long poems, and so on; things they are basically familiar with. I would be interested to see how well they could look at a series of chineese characters and were told to memorize 100 of them and then write them down. I would presume that to anyone who doesn't know chineese, it
  • by Gyan ( 6853 ) on Saturday March 06, 2004 @05:07PM (#8487260)

    In Schacter's memory book [amazon.com], an anecdote is presented about the 1999 National Memory Champion. She commented that she relied on post-its to get through the day.

    It's not really ironic because memory competitions test how transient your memory focus is. Post-Its help those with attentional problems of memory.

    In other words, these memory champions don't have all-around good memory skills.
    • If you think about it, this is testing a person's ability to operate as a tape drive rather than as a hard disk. How long of a sequence can you memorize, rather than how fast you can access small chunks of data after being prompted for them in a not-so-predictable order.

      Ask one of these competitors for just the 87th number in their sequence, and they're going to have to give you the other 86 numbers first whether you need them or not...
    • Seems like there would be a large difference between remembering a few things that matter and a lot of things that don't matter.
  • I've often wondered how the professionals, or even people with more than an average ability to recall do it. I've heard of two different ways. One is to make up a rhyme or a "keyword" to jog your memory of some object, or some series of objects.. The other is to have a snapshot or a visualized picture of something in your head.

    I seem to personally work along the snapshot method, as I suspect many others do. If I close my eyes, I can visualize a page in a text, or a license plate, or a face. Somewhat imperf

    • The Picture/story method is the more common (it's not easy to make up rhymes at once) but the trick is the picture has nothing in common with the object you want to remember!

      Think you want to remember
      in pronounciation:
      R=4
      G,K=6
      S=0
      V=8
      D=1
      So number 46408116 can be represented by sentence:
      RoGeR iS VooDoo DucK

      Now imagine your friend Roger covered with feathers, with lots of pins in his body. Ridiculous? Yes, but hard to forget. And then using the same key as for creating the words, you recover the number from t
  • An informal study of a single website for memory shows that if you are a world memory champion you have a good change of losing the basic ability to formulate English sentences:

    "If you are a already memoriser..."
    (from the front page of the linked website [worldmemor...onship.com])

    Who needs memory when you've got, uh, um, what was it? Dang.
  • by asavage ( 548758 ) on Saturday March 06, 2004 @05:11PM (#8487293)
    I saw a video from the US memory championships from a few years ago. A university professer took a number of fairly average highschool students and trained them using mnemonics for only a few months. These students took 2nd to 10th position only being defeated by a previous champion.

    A summary can be found here [pbs.org]

  • "I wrote them down in my Diary [corky.net] so that I wouldn't have to remember!"
  • How often... (Score:2, Interesting)

    by Orinthe ( 680210 )
    do you actually need to memorize random strings of numbers or letters or positions of cards in a deck or whatnot? Short of trying to memorize 150 digits of 'pi' in middle school for a contest (which was won by someone with "photographic" memory who didn't even look at the numbers until the night before) I can't recall a single time that truly random memorization has been neccessary or useful.

    Instead, our brains are much better suited to recognizing patterns, which is why we can, as actors in play, for
    • Re:How often... (Score:5, Informative)

      by SharpFang ( 651121 ) on Saturday March 06, 2004 @05:17PM (#8487333) Homepage Journal
      > do you actually need to memorize random strings of numbers

      PIN codes, phone numbers, ID numbers, passwords, registration numbers.
      They are hardly ever as long as 150 digits but they are EXTREMELY common.

      > Instead, our brains are much better suited to recognizing patterns

      RTFA, that's what the whole concept is based on. Just associate symbols with patterns/images and then create a story/image based on the set created.
      • Re:How often... (Score:2, Insightful)

        by Orinthe ( 680210 )
        Funny, I have no trouble memorizing PINs, IDs, SSNs, license plates, phone numbers (despite the fact that I hate phones) etc. Only a few of these are actually random (most people change their PIN to a number they'll remember, and even in the case that they don't, that's only a few numbers). Phone numbers are most decidedly not random--they are hierarchical in nature (country code, area code, exchange, and, surprise, only 4 numbers are left to simply memorize)--we have mandatory 10-digit-dialing in my ar
    • you can't juggle? looser. ;)

      I never memorize anything I can look up.
    • My dad used to drive me nuts while playing bridge. He would pick up a newly dealt hand, look at it for about ten seconds (he never sorted it), then put the cards down until the bidding was over.

      At the end of the play he was able to tell exactly which card was played by each player in each trick and if you had made a mistake (God help you), he would tell you which card you should have played instead.

      Then there was the time that he played two games of (timed) blind chess simultaneously against me and my sist

  • by callipygian-showsyst ( 631222 ) on Saturday March 06, 2004 @05:13PM (#8487310) Homepage
    A good memory would help the /. editors: We wouldn't see any more dup articles!
  • A few weeks ago at school there was a competition to see who could memorize the most decimal places of pi. The winner memorized around 160 places I think.
  • by LostCluster ( 625375 ) * on Saturday March 06, 2004 @05:17PM (#8487345)
    Working in trivia as much as I do, I find it interesting how easy it is to convince people that they know something that in fact isn't true.
    One example: They'll read a question too quickly, recall a question they've seen earlier, and then give the answer to the earlier question, not the one that's actually in front of them. They'll then be befuddled why they missed the new question for a while until the actually reread all the words slow enough to see the change.
    • They'll read a question too quickly, recall a question they've seen earlier, and then give the answer to the earlier question, not the one that's actually in front of them.

      I've done that debugging, even to the point of having to narrow it down to one character before I could see it.
  • I remember an episode of Connections that touched on the subject of memory. Specifically the now mostly lost ability to memorize, for example, a lengthy tale or song in one listening and then being able to recite it mostly verbatim. Similarly, the people building the great cathedrals and castles of the past had no blueprints as we know them. The shape and placement of every stone was in the master builder's head. This tends to be supported by the story I heard of the man responsible for building a stron
  • Well, here we have some people people who won't have any trouble remembering their ip addresses when we start using ipv6.
  • by 1iar_parad0x ( 676662 ) on Saturday March 06, 2004 @05:40PM (#8487494)
    "Anyone who considers arithmetical methods of producing random numbers is, of course, in a state of sin."
    -John Von Neumann

    Any decimal number has approximately 3/10 the number of characters of it's binary equivalent. There is no exponential change in the shear number of characters to process.
    [e.g. 2^10=1024, 2^20=1M etc]

    Thus If I encode my data from binary to HEX, I get better "compression" of information.

    Note: IIRC, according to Algorithmic Information Theory, if I were trying to encode "all the data of the universe", then the fact that my compression scheme only reduces the amount of information by a constant and the computation for conversion would probably be so incredibly expensive, there exists no computational gain from Mnemonics.

    However, if I'm given a piece of paper and allowed to use a clever encoding scheme than might be able to "memorize" anything. I only need to memorize a smaller number and the program, which encodes it. Thus deriving my result. Remember, by the rules of this competition I have more time than memory here. Frankly, I think an encoding competition would be more interesting.

    I'm curious as to how this philosophy relates to AIT, Wolfram's Principle of Computational Equivalence, and foundational mathematics.

    "There are two kinds of science -- physics and stamp collecting"
    -Ernest Rutherford
    (Or has he quoted similarly, if I wanted to memorize science, I would have studied botany)
    • by SharpFang ( 651121 ) on Saturday March 06, 2004 @06:03PM (#8487639) Homepage Journal
      But... that IS what they do!
      They encode the data into easily rememberable kind they don't compress it, but rather expand - creating stories, images, pictures, sentences, through the "mnemotechnic memory" technique. Then they decode it just the same way.

      Say, you have a memory medium that can remember arbitrary values from 0 to 256, it has a plenty of room, but it tends to float lightly, i.e. 128 may become 120 or 140 or 100, but not 20 or 210. So for your purpose instead recording byte values, you recode them to binary and record every "1" as two 255's and every "0" as two 0's, then record them. You need 16 bytes of your diskspace to store 1 byte, but it will NOT get lost - only really strong corruption could change the results...
  • Heyyyyyy (Score:4, Funny)

    by (l.windthorst) ( 653814 ) on Saturday March 06, 2004 @05:40PM (#8487495)
    Does anyone remember that Happy Days where the Fonz helps Richie and Potsie study for a Biology test by turning it into a rock song that they play in class? Was that just a weird dream I had? I swear it actually happenned. Well, ever since then I turn things into songs in my head to remember them -- and it works really well. I mean, I added a tune to the Constitution and remembered the entire first two articles. It's kinda scary.
    • It works well. Thanks to They Might Be Giants the facts of James K. Polk's presidency are forever burned into my brain. And whenever someone askes how far the sun is from the earth I can say "about 93 million miles away, that's why it looks so small."
  • Retention? (Score:3, Insightful)

    by -tji ( 139690 ) on Saturday March 06, 2004 @05:43PM (#8487510) Journal
    After reading the examples in the article & on the memory champ's WWW site, the obvious question how well do they retain the memory over time?

    In the competitions, there is a time component. They have a very limited amount of time to commit the information to memory. Then, they must regurgitate it within a short period of time. If they were asked a {day, week, month, year} later, what percentage would be retained?

    Can their techniques be used to retain multiple unrelated data sets simultaneously?

    Basically, the question is: Is this merely a good parlor trick, or a useful mechanism for real-world use?
    • If they were asked a {day, week, month, year} later, what percentage would be retained?

      If they knew they needed to retain it for years they can certainly do so; basically, they would have to revise it at the end of the day, a week later, and a month later; and then they would know it pretty much forever.

      Is this merely a good parlor trick, or a useful mechanism for real-world use?

      Somewhere between the two. They need a mnemonic scheme for each kind of thing they need to learn, but, for example, remember

  • careful (Score:5, Funny)

    by edxwelch ( 600979 ) on Saturday March 06, 2004 @05:52PM (#8487566)
    The human brain has a limited space for memory, so if you try to remeber too much you will end up forgetting other important stuff. That guy that memorized the 3000 long binary number probably can't remember his mother maiden's name now, or where he parked his car.
  • It's doable. (Score:5, Interesting)

    by Mr_Icon ( 124425 ) on Saturday March 06, 2004 @05:55PM (#8487586) Homepage
    Remembering a sequence of 52 cards is actually not that hard. Well, okay, it's hard, but it's doable. I used to be able to do it with relative success, but I haven't practised in over 3 years.

    There are several techniques, and most of them use grouping and storylining. For example, this is the one I used:

    Every card gets three possible meanings -- a subject, an object, and an action. Then you draw the cards in threes and make up a story on the spot. E.g. say you drew a two-hearts, jack-spades, and six-diamonds. In your designation chart, these cards have the following meanings:

    two-hearts: subject: Madonna; action: seduce; object: boobies
    jack-spades: subject: drug dealer; action: wave above one's head menacingly; object: bling-bling
    six-diamonds: subject: bank attendant; action: pay; object: a wrapped packet of dollars.

    So your combination becomes: Madonna menacingly waving a wad of dollars above her head. The key here is to visualize these things and make up a continuous story, as if describing what happened to you on the way to work. (Out of the door, I saw Madonna waving menacingly a wad of dollars above her head. I came to talk to her, and apparently she was angry because a drug dealer shot her car (jack-spades/three-spades/four-diamonds). I offered her a ride, and on the way to her house we saw from the windows of our car Saddam Hussein trying to hump a church building (king-spades/four-hearts/ten-crosses).). It's important to tie the previous action to the next (saw through the windows of our car), so you don't lose the sequence of events.

    The cards are grouped by subjects -- all hearts have to do with sex, all diamonds have to do with money, all spades have to do with criminal element, and all crosses have to do with cults and religion. Usually just three possible meanings per card is not enough, because it can always be that you just CAN'T make something meaningful out of a combination ("Bank teller seducing an electric chair" takes... a lot of imagination to visualise, though if you manage, you'll never forget a six-diamonds/two-hearts/five-spades. Ever).

    Sometimes you sure make up very amusing combinations. E.g. among the ones I recall is Saddam Hussein licking a cash register (king-spades/ace-hearts/ten-diamonds), Marylin Monroe wearing a punctured car tire on her neck (queen-hearts/queen-diamonds/three-spades), and Bill Gates seducing a bill fold (king-diamonds/two-hearts/two-diamonds), though this one could have actually happened for all I know. :)

    The weirder you make your combination, and the more vividly it stands out in your imagination, the higher is the chance that you will remember it.

    Mnemonics is quite amusing. It helped me make it through college without ever taking notes and learn three foreign languages. Definitely a very useful skill to learn and master.
    • Re:It's doable. (Score:2, Insightful)

      by Trailwalker ( 648636 )
      Remembering a sequence of 52 cards is actually not that hard.



      Performed as a matter of routine by any hard core bridge player.


      Easy if you have a reason to remember.
  • by TheNetAvenger ( 624455 ) on Saturday March 06, 2004 @06:01PM (#8487627)
    I do this trick for friends all the time. It is fun with cards...

    You can use it for any serialization of numbers, and cards are very simple. You can also do this with binary (but be good at converting two digit decimal to binary and back).

    Develop a set of references for most two digit numbers that have meaning to you.

    Some I use for example are: 07 - think of James Bond, 22 think of 22 caliber pistol, 13 think of unlucky. It also helps to have a set for single digits, 7 think of lucky for example.

    Then when you look at a series of numbers, all you do is make a story to fit the numbers together.

    For example:
    1307877299220713442

    The story I would make up to remember this:

    Unluckily, James Bond found a RX7 to get away back when I was born. During the getaway, agent 99 shot a 22 pistol at Bond but she was unlucky, and got shot with a 44 magnum twice.

    (The story is often shorter in your head, but I wanted to make it readable for you guys)

    In essence instead of remember numbers, you are remembering the plot to a story.

    Without looking above here is the number set: 1307877299220713442

    13 - Unlucky
    07 - Bond
    87 - Year of RX7 I had a long time ago
    72 - Year I was born
    99 - Agent 99 (from Get Smart)
    22 - 22 pistol
    07 - Bond again
    13 - Unlucky
    44 - 44 Magnum
    2 - Twice

    If you get your associations down for the number pairs you can create little stories and easily remember 100 digit or more sequences of numbers.

    For card tricks, just add color to the story, I use blue and green to denote the difference between hearts and clubs, or sometimes will mix in the heart or spade or club reference into the story (i.e. the Queen took her Spade, etc)

    Most people are impressed if you can just remember the number sequence of a deck of cards and not even bother with the suit, so if the extra colors for the suits throw you, just do the number order of the cards.

    Start with a deck of cards, and I will guarantee you in a few hours or day, you can easily do this.

    Just make up the story as you look through the deck, the faster you know your associations for a story, the faster you can remember the cards. You should be able to remember an entire deck by literally flipping through them as fast as you can read them.

    Happy memorizing... :)
  • I more than doubled the previous first-trial correct score during memory trials at my undergrad, and these were nonsense sentence-pairs (not even normal grammatical structure), where key words of either sentence would be stated, and I'd have to repeat the key words (or the whole sentence) from the other. Two wrong first trial, zero second, zero third (thus, done). Ok, take the morning off!

    The "secret"? Association. For every sentence, I matched words to visual images as they might appear in an episode of C
  • by Dikeman ( 620856 ) on Saturday March 06, 2004 @06:22PM (#8487740) Homepage
    I was checking out the rules for memorizing binary numbers and stumbled over something odd.
    You get 30 min to memorize it, but 60 min to recall it.

    You would think that it would take more time to memorize it, than to recall it. But maybe the speed of recalling is tampered by the speed of writing down 1's and 0's?

    Based on own 'research' i concluded that with normal speed you can write 90-110 1's 0's per minute. The world champion of 2003 had scribbled down 3009 1's and 0's. So that would've taken him between 27-33 minutes. He memorized them in 30 (or less) minutes, meaning this guy can memorize binary numbers faster than i can write them down! But then again, why did he get 60 min to write them down? Do they use special recalling techniques in which you don't continuesly write those numbers?
  • ...
    First, the set up.
    You have to hype up your memory abilities, subtly
    then you start flipping cards over.
    you remeber the first eight and the last 5.
    you bluff the rest.

  • by Pedrito ( 94783 ) on Saturday March 06, 2004 @07:09PM (#8488026)
    Sorry, but it just is. I mean, the way my own memory works just makes no sense. People can tell me things to remember and I just can't. It's not that I don't pay attention when they tell me, it's just after they tell me, it's gone.

    On the other hand, I remember things vividly from as early as 2 years old (events, dreams, etc). I remember phone numbers and lock combinations from childhood (I'm 35 now). Numbers have always been easy for me, though. I see patterns in them and tend to remember the patterns. I have an almost inexhaustable reserve of useless trivial knowledge and God knows why I remember it all. I excel at Jeapordy and Wheel of Fortune. I can understand (read and spoken) 9 languages, but I can only speak 2 of them.

    But ask me to remind you of something in 20 minutes, or tomorrow, or next week, and there's about a 95% chance I'll forget. Ask me what I did 3 days ago and I'm more likely to get it confused with something I did 2 or 4 days ago.

    I consider my memory excellent... For some things. For others, it's just atrocious.
  • by SmackCrackandPot ( 641205 ) on Saturday March 06, 2004 @07:22PM (#8488093)
    Back in the days of the dot com boom in Silicon Valley, you practically had to remember the locations where the doors of the Caltrain carriages were likely to open. This wasn't as difficult as it seems though, since there was a sign indicating to the train driver where he should stop the cab (if not just the train). However, sometimes the train would overshoot, so people would have to frantically run along the platform. In Summertime, parents and schools would reserve the last carriage for children's parties. So people would have to run even further.

    Eventually, we turned this into a casino game: Caltrain Casino

    Each turn was represented by two or three throws of the die/dice.
    The first throw represents which carriage of the train you have chosen. The second throw represents which event has happened. The scoring is as follows:

    [1] Train is completely full and doesn't stop - you lose.
    [2] Last two carriages are reserved for school trip - if you threw a one or two, you lose, otherwise you win.
    [3] The carriage you chose was completely full - If your first throw was three or higher, you lose, otherwise you win.
    [4] Train overshoots. If your first throw was three or less, you lose, otherwise you win.
    [5] Train overshoots by half a carriage. Take another throw. If evens you win, odds you lose.
    [6] Train arrives normally. You win.

    The odds are 50/50 that you will win or lose.
  • This reminds me of The Legend of John Henry [numachi.com] except its memories rather than hammers.

    Could there _be_ a cognitive function of humans that is rendered any more obsolete by technology than edetic memory? Not even arithmetic is as 'rote'.

  • by farrellj ( 563 ) * on Sunday March 07, 2004 @12:14AM (#8489337) Homepage Journal
    One of the specialties of Druids was the Brehons, the judges and law-speakers of Celtic society. They would memorize the entire Canon of Celtic Laws, plus all the precidents that had been decided since the codification of said Laws. Much as the Brits would like us to believe that they invented precident based law and circuit judges, The Celts had such a system in operation over 4,000 years ago. The fact that the British occupied Ireland and tried to destroy Celtic Irish culture is the reason why we don't commonly know the truth of the matter, as the winners tend to write the history books. But some information still exists that allows us to reconize the contribution of the Celtis to both memory, and law.

    Similarly, the Bardic class of Druids memorized their entire songlist, both music and lyrics.

    In fact, because of this memory skill of the Druids, we know little of their rituals and depth of knowledge remain since they memorized it all. And as they were gradually hunted down and killed by the British, Romans and later Christianity, the extent of their knowledge has mostly been lost. We only know what others say about them, for the most part, and one thing that they all agree is that nothing was written down and all was memorized.

    ttyl
    Farrell

"How to make a million dollars: First, get a million dollars." -- Steve Martin

Working...