excreationist

This seems to be about how we recall memories that are appropriate to our current experiences and desires...

I have some thoughts about this - they are mostly based on writings about NLP (neurolinguistic programming) and partly based on computer science/AI and also cognitive science.

Basically I think we have two main types of memory - working memory (that we are directly aware of) and long-term memory. The working memory contains processed sensory data, emotional response data, desires/goals, beliefs/expectations, etc. (some of those categories overlap though) I think the long term memory is a thing that learns patterns found in the working memory and these patterns are associated with other patterns according to the contents of the working memory.

Also, I think the working memory is constantly triggering long-term memories that have some associations with it.

e.g. in our longterm memories, we have the sound "telephone" associated with the string of letters "T-E-L-E-P-H-O-N-E" and the idea that "phone" is an abbreviation and preferred common word for it and what phones look like - what parts they need and how you use phones, etc.

Say in a crossword you had to complete the following word: "T-LE-H-NE"
It is a little tricky because the "h" is part of the "f" sound... but anyway, after a small number of cycles, the working memory would trigger some of the memories in the "telephone" clusters in the brain. These would probably be all over the brain since telephones are partly visual, partly linguistic, partly auditory, etc.
So the sound "telephone!" would probably be triggered along with the entire word...

A more complex example involves creativity...

Say I wanted to think up a creative idea... the problem is that I need to think of a specific creative idea... general examples would include a detailed description of a monster, or a poem, or an invention.

I'll pick poem, and write one, while trying to explain my thought processes (recalling memories, making decisions, etc).

Ok...
Just before I initially thought I'd write a poem about monsters because it means that I wouldn't have to try hard to think of a topic... then I thought I might do something else - this would be because the presence of the "creative" goal in my working memory would mean that immediate ideas, like monsters, aren't good enough.

I think that only useful ideas are triggered from our long-term memory and put into our working memory so that we become aware of them. If you are trying to be creative and are coming up with lots of bad ideas, this is actually useful... these bad ideas are used to refine our search criteria... (it kind of works like a search engine really...)

Since I think the search engine analogy sounds so useful I'll expand on that a bit. (I didn't sense any obvious problems with it)

So say I'm writing a poem, and I need to think up a subject... the first thing on my mind is animals, since I have written about creativity before, using the example of thinking up weird animals... but animals are too boring... I guess my goal includes a subject that isn't very boring. But on the other hand, boring is probably simple. So I'll just go along with animals. So for my first search I did "poem topic" along with the contents of my working memory as stimuli.
The association between creativity and writing an example (the main focus of my working memory) and [weird] animals was very strong, and this search engine returns the strongest associations first.
Since I wanted a certain level of certainty in my results, I automatically did another search "are animals a good enough subject for a poem?"
Now that I think of it, I think I have a fairly strong association between most poems, such as animal poems and boredom. So that was a relevant problem so the concept of boredom in association with animal poems was triggered.
So I considered that perhaps having animals as a subject was a bad idea... but since time isn't much of an issue, I wanted to be sure, so I did a follow up search: "are animal poems really that bad?" and I think I also triggered another search at the same time "what subject could I do instead?" and the first search finished with a result first and this result was that it doesn't really matter, based on some things such as the length of this post getting longer (which should be avoided) and that it doesn't matter if my poem is boring.

Anyway, after that I had some thoughts about which animal... my first thoughts were cat... dog... This would be because I associate "animal" with those words the strongest and I am in close proximity to my cat and dog. If I was at a zoo I'd probably say something else - like monkey.
Anyway, I wanted to be fairly sure of my answer and I found quickly that cat and dog aren't good enough. Then I came up with "sheep"... maybe it's because a lot of children's poems and books involve sheep. I guess I had a strong association between animals and poems and children's poems. And a strong association between sheep and children's poems. I wasn't immediately aware of my chain of associations - since it was irrelevant... I was only immediately informed of the final result of the search.

The length of this post was on my mind the whole time and basically I was subconsciously thinking that longer is bad, and the poem can be scrapped, so I'll forget about writing a poem.

Here is how my reasoning would go:
-> means "triggered"

post getting longer -> long post bad

(post getting longer, long post bad)
-> (x getting worse therefore prevent x)

Apply template [x getting worse therefore prevent x] to post getting longer -> stop <post getting longer>

stop <post getting longer> -> stop one of the necessary causes

the action "typing" is associated with the result "post getting longer"

"a necessary cause" is the pattern of one thing (A) consistently being associated with another (B) where there are two processes and one (A)preceeds the other (B). The association might only be very strong one way though... "A" mightn't guarantee "B", but "B" would require "A".

Anyway, subconsciously I'd ask myself how I could prevent the typing and continued appearance of letters on the screen and I'd eventually come to the conclusion that I'd need to end the post soon by finishing what I have to say and submitting my reply.

BTW, the processes I described here are very, very rough approximations to how I think our reasoning and recall works... but hopefully people will get the gist of it.

Info about <a href="http://www.timestocome.com/javaai/nnassocmemory.html" target="_blank">Associative Memories</a>:
<a href="http://www.comp.nus.edu.sg/~pris/AssociativeMemory/AssociativeMemoryContent.html" target="_blank">more info</a>:
That means that BAM's can be used to do things like associate people's names with phone numbers - so you could give it the phone number and it would recall the name - and vice-versa. They do the search in parallel though rather than one name or phone number being analysed at a time (like how a computer typically does it).

Ierrellus

sweet as a nut,

Mea culpa. It was probably some other device that showed the dream state in the brain, not positron emission tomography. When I posted that, the vivid memory in my mind was that of fireworks blasting from the back of the brain.

My short term memory often fails because of my illness. Once I got out of my car, threw my car keys to the ground and tried to "put them out" while I was about to stuff the cigarette into my pocket. Luckily, my mind says, "Something is wrong with this picture!" To me, the incident was funny.

I also have to watch while writing because I come up with portmanteaus that Lewis Carroll would have been proud of. Enjoying the good posts on this thread, I have to reread them often to remember them. Life is interesting!

Ierrellus
PAX

snatchbalance

excreationist,

Do you have any thoughts on, what might be termed, "long term search"? I'm refering to the phenomenon where one will "forget" about a problem, then the answer will appear, sometimes months later, spontaneously, maybe even in a dream.

The dredging and processing of memory seems to be taking place completely in the background.

sb

sweep

hey, were only human ierrellus so i like to make mistakes, that's partly what i am here for: so that everyone else can smash my model to bits so that i have to build another - 'bigger, faster, stronger'

tell me a bit more about your illness, now i am hooked- i hope your lungs aren't too mashed from all the smoking. I would like to hear some 'portmanteau'- isn't that something to do with schizophrenia?

sweep

tron, sorry&gt; if you look at the whole of my message you'll see that it is sometimes easy for me to see things and be blind to the same elements moments later, although i see you fall into the same trap as i do. I won't make this an issue, so please don't be mad at my picky-ness.

but....

compared with:

i suppose you mean by remember we bring to light, we think about. i agree we model memories but when we remember them, are we not also reconstructing them? :- as in "my tower needs to be renovated, a bit of plastering, repointing, strengthen the foundations, a few embellishments"

another thing, i can see that we don't get to choose what comes up first when 'i will' but i can choose to think and not to think, or to block thoughts out. What does the blocking and the choosing? i suppose we do have a mechanism for this task, but i don't expect you to answer me, only that you acknowledge the problem at hand.

regards

[ August 14, 2002: Message edited by: sweet as a nut ]</p>

sweep

excreationist, thanks for the input. that is a very interesting account and i have come across those notions before. Does the working memory monitor what is going on in the brain?

in the meantime, steaming, grey miasma is oozing from my ears. Please excuse me while i go and clean the carpet.

[ August 14, 2002: Message edited by: sweet as a nut ]</p>

sweep

from DRFSeven:

hmmm..

snatchbalance

Sweet as a nut,

I think dfr7 answered your question. You need to buy her a pint.

sb

sweep

I would be delighted to: what a star!

Bill

You know, your robot analogy isn't as far off as you might think that it is.

I'm in the process of reading this month's II Book-of-the-Month, <a href="http://www.secweb.org/bookstore/bookdetail.asp?BookID=836" target="_blank">Religion Explained</a>. In that book, Dr. Boyer explains that the human brain is actually a complex collection of "inference engines," each operating in parallel but separate from all of the rest. If your analogy of robots is equated to Boyer's "inference engines," then things might really be closer to what you had in mind in the first place.

I believe it is well accepted "that information is stored in the brain." While information is certainly stored in other locations (such as in our genes), the brain is the central repository of information that we can access with thought.

However, accessing that information isn't as straight-forward as you might first believe. You don't have memory cells like pigeon-hole boxes that hold information. Instead, memory works more holographically than that, storing (literally) little bits and pieces of the required data in various spots within the brain, making up rather a complex structure. The selection and production of information is not done directly, but is rather done through the operation of inference engines. In other words, you don't remember the picture of your car, but you rather infer what your car looks like by remembering a complex collection of data about the car and assembling it "on demand" when you need to recall what your car looks like (such as when you are walking through a parking lot, looking for the once car that screams "mine" at you). This sort of analysis is produced through the operation of Boyer's "inference engines."

I hope that this observation adds something to your understanding of the real answer to the question you first posed....

== Bill