Re: Infinite Memory: Question 17

From: Harnad, Stevan (
Date: Wed Apr 30 1997 - 18:45:33 BST

> From: Gray, James <>
> We do not remember everything because:
> A. our memories are not good enough.
> B. there are some things we just want to forget.
> C. our heads are not big enough.
> D. it would be too confusing.
> E. none of the above.
> I personally think that C & D are the answers as if we could
> remember everything we'd have to sift through loads of rubbish to find
> the information we wanted, this would slow processing time down
> dramatically until really old people would just seize (not that they
> don't now) from excess knowledge.
> If we tried to remember everything our brains would need huge amounts
> of storage space and our heads wouldn't be big enough! Go fig.

Well, that's not the right answer, but since so many of you were thrown
by this one, I won't count it.

Here is a general suggestion about exams and quizzes: It's certain that
they are not testing ordinary everyday knowledge. So if the answer is
one that you could have given from never having taken the course, then
chances are very good that it's wrong.

The correct answer is E, none of the above.

If you want to know why, see page 91 of the book, and the skywriting
notes on that chapter (4).

Here's an excerpt:

    "Remember that the shadow an object casts on your retina changes
    shape depending on the position and distance of the object. How do
    we know it's the same object? Here's one way we DON'T do it: By
    memorising every single view we ever have of it:

    "The novelist, Borges, wrote a short-story, "Funes the Memorious",
    about someone who could remember everything: every instant of his
    life was exactly recorded in his mind forever. That may sound like
    an advantage, but actually it was a huge handicap -- so huge that a
    man like Funes could only exist in fiction. Funes could give a
    different name to every number as far as he could count -- 1 would
    be, say, "Pedro," 2 would be "Luis," 1,246,937 would be Jorge, etc.
    -- but he couldn't do calculations with the numbers (how could you
    even add them if each one was unique and had only its own proper

    "Funes also had trouble with calling (what we would call) "the dog,
    Rover," by the same name when it was in two different locations at
    two different times, because for him every instant was unique and
    remembered for ever. He could give a proper name to each "snapshot"
    instant in his life, just as he could name each number, but he
    couldn't find the INVARIANTS in those snapshots -- the things that
    stayed the same from snapshot to snapshot."

> From: "Anonymous" <>
> In regards to our not having enough "storage space", I once read that,
> supposing it were possible to consciously remember everything, we could
> learn a million things every day for the next million years and still
> not fill our memory up. In my opinion, the storage space you find on
> computers pales in comparison next to awesome memory capacity we have.

Interesting, but do you actual know any evidence about that?
"I once read" is not evidence...

> I believe there are also some theories suggesting that we do
> sub-consciously remember everything and that there is some sort of
> filter mechanism that prevents us accessing it all. I think that this
> may be the basis on which techniques such as hypnotic regression work,
> i.e. by-passing this filter. Though of course these techniques are also
> capable of implanting false memories but that is a different kettle of
> fish.

In Year 3's Current Debates course you'll hear about controversial
issues such as hypnotically induced memories. But don't be too ready to
believe things you hear about memory, hypnosis, and the alleged
"untapped potential" of the brain...

> From: "Anonymous" <>
> The Frame Problem exists because we cannot programme a system with all
> the rules it needs and, even if we could, it would be the size of the
> planet. Or something.

Actually, the frame problem does not arise because of limited storage
capacity or computational resources, because both people and computers
have limited capacity and resources. It arises because knowledge is not
just symbols and symbol manipulation rules. Any system that has just
symbols and symbol manipulation rules runs into the frame problem,
because the symbols and rules have not anticipated every possibility.
So even when you fix one frame problem by putting in some more symbols
or rules to take care of that case, another frame problem is always
waiting around the corner.

> So, if we were talking about algorithms and artificial symbol systems
> our heads would not be big enough.

No, the reason we can't remember everything is not the same as the
reason symbol systems keep encountering the frame problem. For one
thing, WE don't encounter the frame problem, only symbol systems do (so
we're not just symbol systems).

> But that's not what the question is about.
> Is there an answer to this? Why would we need to remember everything?

Well, we've all mused about how things would go better if we could
remember more. So it seems reasonable to ask why we don't remember more
-- why not remember everything? And although the layman's idea about
this is that it's because our heads are not big enough, the real reason
turns out to be that you need to be able to remember what is invariant
across a lot of different instants of experience in order to be able to
do anything at all: learn, classify, generalise, predict, etc. --
even to be able to name things, and kinds of things.

This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:52 GMT