Beefy Boxes and Bandwidth Generously Provided by pair Networks
Syntactic Confectionery Delight

How do our brains work?

by japhy (Canon)
on Oct 17, 2000 at 19:56 UTC ( #37162=perlmeditation: print w/replies, xml ) Need Help??

I feel my brain is a hash of hashes (of hashes, etc.). I do not think arrays (as such) enter into the picture, really. I remember things via other things, not by when I learned them, per se. And I remember sequential things via linked lists (which COULD be seen as either:
$alphabet = { letter => 'a', next => { letter => 'b', ... } };
or as
$alphabet = { a => 'b', b => 'c', c => 'd', ... };
or any other multitude of ways). This implies that I think in a relational sense -- I best remember things by WHERE I've seen them. When I had to memorize the founding fathers of my fraternity (Acacia), I remembered which came after the next -- I could not say "the 8th father is ...", because I didn't have array-indexing abilities. I had to step through the list to find the 8th.

One of the brothers I just asked says he uses associative (which is compareable with relational) thinking, and that he uses linked lists for remembering sequential things. The argument he used is "when you smell a flower, you remember happy things, right?" That's associating the odor with a pleasant thing.

Then something interesting came up:
  • Jeff: Most people can count backwards because numbers have a repeating pattern, but they can't say the alphabet backwards because there's not a pattern to the letters, it's just "b" comes after "a", etc.
  • Will: No, letters have a base 26 pattern than just doesn't get a chance to repeat.
  • Jeff: Damn.
(Maybe that was just a fun tangent, I dunno.) So I'm curious how you think your brain works in the Perl sense.


Replies are listed 'Best First'.
RE: How do our brains work?
by princepawn (Parson) on Oct 18, 2000 at 00:02 UTC
    Hmm, well I have a Master's Degree in something called Computational Neuroscience, an enterprise devoted to looking at the nervous system as a computational device.

    Regarding your statements about associative memory, if the title of your post limited the discussion to the memory retrieval function of the brain, then your assertion would then be 1/2 OK... 1/2OK because there are two types of memory: declarative and procedural. Declarative memory is: "my name is bob" Procedural memory is how you remember how to throw darts --- the set of procedures to do something. What you (and all other posts) were discussing was declarative memory.

    One neat thing about the brain is fault tolerance, meaning, that unlike a computer program, it can still manage to perform tasks in spite of input or self degradation. That is to say, a Perl program won't run at all once it is missing one semicolon and all searches for Purl will fail all associative search engines, but any human would be able to deal with the poor input and still make the relation after that night of drunken-ness.

    Also, memory plays important part in other brain tasks. Take Neural Robotics, where robots perform various tasks (reaching and grasping, walking, etc.) based understandings of how mammals do it. Because the more times you do it, the better you get. But another thing that the brain is doing during this task is motor refinement via visual feedback. Meaning, you are looking at the act and based on what your brain "sees", the brain then refines the motor commands. And the motor commands are largely just increases or decreases in excitation to the muscles. This is brain controlled, but certainly has very little to do with association.

    Finally, most of the best programs which attempt to simulate brain function have had to make serious re-considerations based on real physiological evidence from the brain. Both biological and artificial neural networks up until last year both used fixed numbers of inputs to the neurons and attempted to store memory in weights. The artificial weights were simple numbers and the biological weights were more complex and biochemically based. But in both cases, they assumed fixed connections between cells. But G.G. Turrigiano of UCSD, in Thinking globally, acting locally: AMPA receptor turnover and synaptic strength. Neuron. 1998 Nov;21(5):933-5 (you have to pay for online access to this paper --- no href provided) showed that no brain cell has the same synapses permanently. This means that even though the brain has relatively constant memory store, the memory can't be in the synapses because the receptors are constantly turning over.

RE: How do our brains work?
by KM (Priest) on Oct 17, 2000 at 20:57 UTC
    Memory is associative. Our brains work more like a large relational database. I say database because a database is more permanent than a hash, and is still curruptable :) You can also look at this as a large HoHoHoHxN datastructure. For example, this is a way a brain can make relations:


    You can see, all this information in linked together, and various links share common items. This is going to all be part of the neural network which links items together to form your memory (like a good relational db may use keys and indexes), and assist with recall.

    because I didn't have array-indexing abilities. I had to step through the list to find the 8th

    I also had to remember mine for my fraternity (TKE) a few years ago. I had to remember which was the 8th, as well as who he came before and after in the lineage. So, my memory structure looked something like:

    %founders = (1 => {Name => "John Dough", Preceeds => "Jack Mehoff", Af +ter => "I.P. Freely"}, 2 => etc... )

    Now, I built relations between numbers, names, and numbers and names. When asked who is number 8, my brain had a few ways to connect the information:

    7s Name-8-Name

    Then, after a night of heavy drinking, I would likely loose one or two of those relations by killing brain cells (a random delete on the hash, or database corruption), but could still recall the information because I had various relationships.

    Will: No, letters have a base 26 pattern than just doesn't get a chance to repeat.

    Not only that, but we are conditioned to learn the alphabet forwards. If you build the relations of the alphabet in reverse, you can do this. This is sort of like asking someone to say the alphabet without singing the alphabet song... most (at least in the US) learned with this song, and the tune is related to the alphabet. I also had to do this for my fraternity, but it was the Greek alphabet.. forwards and backwards.

    A phone number may be remembered sequentially. When we remember phone numbers, like 888-555-6473, we practice the pattern in sequence. I wouldn't tell you what number comes before the 7 without starting from the beginning. The numbers aren't related to eachother in a truely relational way (unless you practice it), only to who owns that number, and such. Generally, people put phone numbers into short-term memory (a temp hash), and have to put it into long-term memory (the db) by relating the number to something(s). But, when you remember something like this in a list, you will need to find meaningful ways to associate one number from the others around it. If you chunk the numbers together from the end, you will likely achieve this.

    Human memory is very complex, and doesn't simply involve something like hash key lookups, because links can be broken and things can be "forgotten". This is retrieval. Retrieval can work really well in a soundex fashion (as well as remembering things in chunks). For example, studies have shown that classes over 45 minutes give students chunks of information which are too large to remember. Or when you think "Oh! The name of that song is something like 'Shallow Hole'.. er.. 'Mallow Pole'.. oh yeah, 'Gallows Pole'". You scan your memory for things that are similar, until you make the correct association. Ok, wait.. I am getting too much into how Human Memory works (grad school flashbacks).

    So I'm curious how you think your brain works in the Perl sense.

    In a Perl sense? Well, I think it would be large HoHs which would be short term memory. To store it in long-term, I would use DBI to stuff the data in useful chunks and relations into a database. To retrieve the information, I would use a large SQL statement with a large WHERE clause :)

    select memory where key=flower and environment=outside and result=snee +ze

    That will recall a memory that the wind will carry pollen of a blossomed flower and make me sneeze. I would hate to be the DBA who needs to design a database to replicate human memory and recall!!


      After careful consideration of your and japhy's posts, I can only conclude that I'm glad I wasn't in a fraternity. :)
      I'd have to agree with associative thinking as a start. The chains read perfectly to me. Following the "flower-plant-water-grow-bloom-seeds" example, each one builds not only on the proceeding word, but also the entire list. Just as "flower" and "water" had different next terms, they stayed within context. If I was given "water" as a starting point without "flower-outdoor" or "flower-plant" as a context, there are many more choices then just "grow" or "bloom".

      As such, the entire context is taken, not just a simple linked list. Much uglier to code. 8) Each member of the list build on the entire connotation to focus. Any one item is virtually unlimited in potential associations. If we thought context-free, then every time we started thinking, we would be lost in a wonderland until we defined enough of what we were thinking about to give us roadmarks and help narrow our path to something we can deal with. In most cases, this is not the situation.

      One aspect of this could be non-sequitors. They could be associations that fall outside the normal context, so they allow a large range of connections until focus is achieved. Any ideas?

      Play word association in a social setting. Then play by e-mail where you only send on the current word. Look at the whole game afterwards. There should be a large difference in the flow of game, when in the former everyone heres the context, and in the later people only have the previous word. And this still isn't context free - you also have all of the previous "turns" that you've had a word come to you and submitted a word yourself. (Anyone want to code a quick CGI to try this with?)

      =Blue might be eaten by a grue...

      > Not only that, but we are conditioned to learn the alphabet forwards.

      There was a schoolyard folklore when I was a kid, that if you ever successfully repeated the alphabet backwards, The Devil(tm) would appear... or was that the Lord's Prayer?

      > When we remember phone numbers, like 888-555-6473, we practice the pattern in sequence.

      "Me too, I ate one sour too." - Fat Freddy

      The phone numbers I remember best, I don't store as a sequence of numbers, but rather as the pattern on the keypad. Strange, no?

         - Muse (who is always reminded of San Antonio when she smells acacia)

        The telephone company (the *real* telephone company, before that Communist b******d, judge Harold H. Greene, got his politically motivated mitts on it) spent a fair amount of money determing how people remembered phone numbers.

        Back in the days of rotary phones (I doubt but a few of you are old enough to have used them. Perhaps you've seen them in museums and such), when phone numbers were shorter, they used a couple of letters, and the digits, like 'BR-549', or some acronym for the letters ('Slick Willy-666' for SW-666).

        When semi-modern (non-VoIP) phones (ones with DTMF pads) were developed, and a standard length number format was settled on, they spent much more money learning that people grouped digits to remember them, and the 3/3/4 method of expressing a phone number came about. Along with this study came the layout of the dialing pad itself, with 1 at the top left. This came out slightly ahead of another possible payout, more like a common calculator would later use. It was determined that for the purposes of remembering the numbers, the TL1 (TopLeft1) layout aided memory better than then the BL1 (BottomLeft1) layout.

        For some reason, the Europeans never took this to heart. Since it was a lot harder to get various countries, each with their own switching equipment ranging from pre-crossbar-1A systems up to the latest switches, to all upgrade together, the European phone numbers range from 4 to 13 digits, sometimes even within the same country. Imagine trying to work with that...


        e-mail jcwren
        My sister noticed that one of her beauty creams has acacia in it. That's cool. My fraternity gets around. ;)

        As for phone numbers, it depends. I've been able to memorize serial numbers on money and credit card numbers and driver's license numbers by finding patterns. I do that for some phone numbers too, but I especially like it when the phone number offers a visual pattern on the keys. This is made far simpler by the fact that the keys are in a square (and then 0, of course), unlike the numbers at the top of your keyboard, which are in a single row.

      I also had to do this for my fraternity, but it was the Greek alphabet.. forwards and backwards.

      You forget to tell 'em about having to hold a lit match while you recite . . .

      Heh . . . geek hazing:
      having to count from 0 to 30 in hex
      backwards while holding a lit match

RE: How do our brains work?
by swiftone (Curate) on Oct 17, 2000 at 19:59 UTC
    how you think your brain works in the Perl sense.

    It doesn't. What's closest? Probably some form of non-binary tree. I can only recall some things after following certain mental routes ("Where did you see it last?").

    But I don't think in terms of lists, arrays, or hashes. The brain is much closer to a network than an script.

      That would be associative, then. Let's say you were sitting at your desk, programming, and you thought of something you wanted to buy -- it was in your short term memory, and then you forgot it. It wasn't until you remembered you were programming when you thought of this object that you remembered what the object WAS.
      %brain = ( programming => { 'things to buy' => { 'doohickey' => undef, }, # ..., }, # ..., );
      That's how I see my brain working.

        Well, your brain would make more connections than that one:

        %brain = ( programming => { 'things to buy' => { 'doohickey' => undef, }, # ..., }, things_to_buy => {'milk' => '1 gallon', 'eggs' => {'amount' => 'dozen', 'color' => 'brown', 'size' => 'Large grade A' } 'doohickey' => undef }, doohickey => (\$brain{programming}, \$brain{things_to_buy}), etc... # ..., );

        So, you would recall it in various ways. You may be at the store buying milk, you may be programming again etc... So, not that I disagree here, but the brain is more complex than any simple hash structure could illustrate.. it has triggers which will fill in other relations, according to various relations you have already made, and create new ones, remove some, blah blah blah.


        Yes, if you wanted to represent that in Perl, that'd be the closest thing. But How do you express "This math solution _looks_ wrong", or the times that something you were thinking about something and gave up, and the answer comes to you out of the blue two days later mid-beer? Or how you have to navigate all of these associatives to get to something, but the next day it's the first thing you think of?

        You can express thoughts in Perl, and it's even kind of useful, but that doesn't mean it represents how you actually _think_.

RE: How do our brains work?
by tilly (Archbishop) on Oct 17, 2000 at 20:25 UTC
RE: How do our brains work?
by OzzyOsbourne (Chaplain) on Oct 18, 2000 at 15:53 UTC

    "No system can fully understand itself."

    It applies to computers, AI, and, humans. The brain cannot determine how it thinks because the act of thinking actually affects what is being analyzed. Plus, there is no way that the part that is thinking can analyze itself because it it being used.

    To think about this on an easier level, apply it to computers: How can the thread that is doing the analyzing be analyzed at the same time? With another thread? You are then affecting the system by affecting processing percentage and throwing off the system analysis.

    Or so they say...

RE: How do our brains work?
by isotope (Deacon) on Oct 17, 2000 at 21:46 UTC
    Just another thought... our brains don't just store strings, or numbers, but also video, audio, all the other senses, and abstracts (feelings, moods). I think the data structure would be more like a three-dimensional mesh, with various processes (dreams, for example) dancing around creating new interconnections between elements.
    What I don't understand so well is how I can read a chapter of a book very slowly and methodically and not remember much more than vague ideas 10 minutes later, but if I speed-read the same amount of text (about 10 pages per minute), I actually retain much more of it (short term, anyway -- I did it all the time for a particular class that required a lot of outside reading for in-class discussion).

      Dreams do more than add new connections. They use older patterns to reinforce newly learned behavior. Part of how you remember and learn, especially procedural memory, is repetition. Plenty of studies exist showing that you need to repeat things to get it to stick. Thus the make-you-do-it-over-and-over nature of some school work, the chanting of product names in 15 second commercials and more. The average number of repetitions needed to get you to stick it in longer term memory? _9_

      The things that have been shown to get the same results with fewer repetitions? Humor can cut the number to 6ish, eating while learning, can cut the number to 5ish (thus primetime, just after people have eaten, and newstime, as they are eating are 1-2 in ad dollars, with number 3 being morning before drivetime, breakfast. It isn't just that you are in front of the TV, you are more vulnerable while eating.), pain can cut the number to 1 =), dreaming can cut the number to 3. Bedtime stories are among the best remembered stories in the world.

      The problem with reading slow vs. speed reading is one of focus, when going slow and thinking, your thoughts on the matter get muddled in with the facts. OTOH, conclusions you come to while reading slow will still be with you long after the facts have drained away, change your strategy based on what you need, the lesson or the facts.

      $you = new YOU;
      honk() if $you->love(perl)

      Actually, dreaming would be more analagous to a garbage-collection routine, as it throws out unused variables from Memory::ShortTerm and moves some to Memory::LongTerm.

        Aside from the garbage collection, I do notice that my dreams help me create strong relationships between data. One of my study strategies was to read through all my notes and coursework one time right before going to sleep. I would dream about the information, and when I awoke, I had a very thorough understanding of it. It worked great for classes where the professor tested us on *everything*.

        Actually dreaming would appear to be far more complex than that. Try this for some interesting thoughts.

Log In?

What's my password?
Create A New User
Domain Nodelet?
Node Status?
node history
Node Type: perlmeditation [id://37162]
Approved by root
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others having an uproarious good time at the Monastery: (4)
As of 2022-12-02 05:48 GMT
Find Nodes?
    Voting Booth?