Tuesday, November 27, 2007

Recall

i'm sure you've noticed that you recall some things much easier than others. At least i do. Maybe you recall everything perfectly. If so, remind me never to play Trivial Pursuit with you. If you're like me, though, you depend on some combination of association and indexing for most of your recall needs.

Most of us are familiar with free association games. If you are asked to say the first word that pops into your mind when hearing the word red, you might say apple. i might think fire truck. Someone else might respond with 650 nm (if you ever play this game with Techers, be prepared for some really random associations). Even better is playing non-sequitur, where you have to say a word that is not related to the word just said by someone else. It's actually very hard to keep up, especially if you're trying to be fast, and reveals the power associations have in our brains.

<grossGeneralization apologiesTo="Dan" >Our brains are made up of billions of neurons, each with many connections to other neurons via synapses. When a neuron is activated, it communicates its activation to its synaptically connected neurons, which may or may not activate them as well. The more a certain activation pathway is used, the more likely that it will be activated in the future.</grossGeneralization> i like to think that association works the same way, but on a larger scale (i think that makes it homologous, but i'm not absolutely sure. Maybe analagous is a better term?). On this scale, your brain is made up of a bunch of different words or concepts, with connections of varying strength to other words or concepts. So when you hear the word house, you might think of a door and see an image of your own house, which reminds you of other houses you have lived in, and so on down rabbit trails and off on tangents. i think that tangents, then, could be explained as particularly strong associations down which you are easily led (or just enjoy following). Some humor, as well, could be defined as finding somewhat obscure associations and raising our awareness of them. As with neurons, these connections between words/concepts grow stronger the more we use them together and weaken over time if not used.

Personally, i find that i do free association very well, particularly in the context of multiple choice questions. It's a big part of the reason i do well on standardized tests (honestly not that great of a skill - i don't have standardized tests at work, and Gavin didn't come with one either). Over time and repeated practice, i have learned to 'trust my gut' when faced with a question that i might not know a lot about. As long as i have run across the information asked for, it is possible that i will lean towards the correct answer. This is by no means infallible, but i tend to do better than random, even when i don't really know anything about the subject. i just have these faint associations calling out, "Pick me!" The more information given, the better we tend to do (in general), because we have a greater number of possible paths to light up the correct answer (as long as our initial associations are good - it's easy to pick up bad associations). That isn't always true, though, because sometimes the information might be given in order to lead us off the mark. Framing a question differently can often lead to different answers. A Pavlovian response is also the product of a strong association that may not be based in any rational method. When i hear a fire alarm, i think doughnuts. i can feel the tangents pulling me all over the place. Must. Stay. On. Track.

Indexing is a much less commonly used idiom for memory recall. It might not really be used at all, but i'm going to manipulate the term until it means what i say it means. When i say indexing, i think of databases (there goes that association thing again). One of the most common ways to speed up the response time of a database is to use indexes. <grossGeneralization apologiesTo="DBAs, anyone who actually reads this" >When a database table is created, it doesn't know how you want to search it. For example, i can create a table called Friends, and it has fields for first_name, last_name, and coolness (no, i don't rate my friends like this, it's a much more complex algorithm). When i want to find all friends with a coolness greater than 5, the database has to go through each record in the table to compile my list. Alternatively, i can tell the database to index the coolness field. The database will then keep a separate structure of the coolness values, along with which rows they relate to. It gets a bit complex in implementation, but the basic idea is that searching that separate structure is much faster than searching the entire table. </grossGeneralization>

i think that our brains index as well. As an example, think of the rooms in your house. You likely can come up with a comprehensive list fairly quickly. Yes, there is association going on here. Maybe it's all association under the hood. But when coming up with lists, some constraints make it easier than others. It's easy to think of boy names that begin with B - Brian, Bill, Bob, Bart, ... But try to come up with a list of boy names where R is the second letter (admittedly a smaller pool) - Arthur, Brian (by looking at my previous list), and i quickly degenerate into trying random sets of syllables (Dray... Draw... Drew - that's a name!). i can feel my brain stuttering. i can't constrain my search set well, so it feels like i need to examine each name out of all boy names to see if the second letter is R. i simply don't index names by second letter. i suppose you could say i don't have strong second-letter associations, but i find it slightly easier to think about it as indexes. i think a lot of memorization falls under the indexed category - the strongly associated set. Presidents of the USA. Noble gases. Elementary particles.

So, let's outline some different questions that apply to association and indexing to different degrees:

Free Association:
Question: Name 5 words you associate with green
Difficulty: easy

  • Grass
  • Oscar the Grouch
  • Go (traffic signal)
  • Felt
  • Gavin's Bumbo
A lot of variation here


Constrained Free Association:
Question: Name 5 blue things from college
Difficulty: medium

  • Blue Books
  • Tommy
  • Merzbacher
  • Jackson
  • The paper around those white erasers
Easier once i came up with a theme (textbooks - basically turning it partially into an index)


Non-Sequitur:
Question: Name 5 things that aren't associated with movies
Difficulty: medium-hard (mental thrashing)

  • Computer monitors
  • Sudoku
  • Log cabins
  • Leap frog
  • Gardening
It's hard to come up with things that aren't associated in some way. i'm not satisfied with the list i came up with, and i ended up not using a lot of the ideas that i did think of. It's so easy to come up with association paths ( sudoku -> crosswords -> didn't they make a movie about crossword aficianados? If not, then -> scrabble, and i know they made a movie about big scrabble tournaments).


Simple Indexing:
Question: Name 5 authors in your bookcase
Difficulty: easy

  • Isaac Asimov
  • Orson Scott Card
  • David Eddings
  • Frank Herbert
  • JRR Tolkien
For this one, i almost look through my shelves to make the list. i know more or less where the particular authors are located, and it's basically an exercise in 'looking' at the shelves in my mind.


Bad Indexing:
Question: Name 5 book titles that include the word Free
Difficulty: hard-impossible

  • ?
  • ?
  • ?
  • ?
  • ?
i am pretty much terrible at this sort of thing. Doesn't work for me at all.

4 unique comments:

Adam said...

Wow, I had such different associations:

Green:
Green Grass
Green Plants
Green Beans
Gasoline (as in not green fuel)
Mowing

Blue from college:
Blue Sky of our Loft
The concoction that Stephen and Nate made oscillate from Blue to Red
Blue flame of acetone
Blue Powerade
The A button on the N64 Controller

Not associated with movies:
Sandals
Salami
Tennis
Legos
Wonderbread

Authors:
Orson Scott Card
Kurt Vonnegut
CS Lewis
Lewis Caroll
Frank Peretti

Bad indexing:
Freakonmics is all I could come up with. But I don't think that counts.

Thanks for the brain exercise.

Anonymous said...

It's a bit strange to realize that you and I think so very similarly. Sure, a lot of this comes from a common frame (we both frame a part of the world in terms of Tech), but I think it's more than that. I also do relational thinking well, do well with standardized tests, I try making up words until something sounds right, and I mentally associate HTML tags with my sentences.

That being said:
Green:
grass
money
envy
an eye color
turtles

Blue from college:
Blue books (dur)
The sky, when it wasn't hazy
Jackson
the color Molly and I repainted around 5A senior year
gene pool

Not associated with movies:
nostrils
TV writers (ha!)
MIT
tapirs
socks

Authors:
Gaiman
Tolkien
Rowling
Brian K. Vaughn (comic author)
Rushdie


Bad indexing:
Uh ...
Play freebird?
Free your mind?
Born on the Fourth of Free-ly?
... Nope, I got nothing

chad said...

Free might be on the hard end of words included in book titles. How about king? i can think of a few.

Unknown said...

Born Free