Re: On Remembering Everything (by DC)

From: Stevan Harnad (harnad@cogsci.soton.ac.uk)
Date: Sat Mar 02 1996 - 17:18:51 GMT


> From: "DONNA CRUMLEY" <DHC195@psy.soton.ac.uk>
> Date: Sun, 18 Feb 1995 20:37:26 GMT
>
> Why did these people forget life events which only happened in the
> previous year? One idea is that as time passes the traces left by
> outside events in the nervous system grow fainter until they can no
> longer be reactivated.

But why do we forget? Just because we don't have the capacity to
remember everything? Is it just a storage size question? Or is
some forgetting needed for some other reason?

> his images were so vividly remembered that they don't
> cancel one another out and so they interfere.

This is beginning to point to an answer, but on the face of it, as long
as there are discrete, distinct "instants," if each could be remembered
fully, there should be no "interference": each instant is unique. The
trouble is that S is asked to selectively remember items; those items
have to be picked out and named. It follows that other items of the
experience must not be picked out. It is in this selectivity that
infinite memory becomes a problem: Being selective means ignoring,
suppressing some parts of the memory in favour of others. Infinite rote
memory means remembering all of it. This is more a matter of conflict
than interference: How can you remember everything and ignore some of it
at the same time?

> This problem of not being
> able to understand science is also to do with his defect in cognition.
> His infinitely faithful memory for perceptual detail made him miss
> patterns and regularities across different events: Each was so well
> remembered that S had problems abstracting regularities and ignoring
> relevant details. Therefore science proved to be a problem because it
> is mostly about abstracting regularities and ignoring irrelevant
> details.

Not just science (what is science?) -- anything that involves naming and
categorising, for this always requires selectively ignoring some things
and not others. Strictly speaking, if S or Funes really had 100% recall
for every instant, then they could never name anything at all, hence
could not speak about anything. Nor could they respond selectively to
certain kinds of things (e.g., food), so they could not survive. Rote,
concrete memory is merely a sensorimotor videotape; it's fine for
replaying scenes, but not for DOING anything with or about them.

> he had a '
> poor memory' for faces; because he saw faces as changing patterns of
> light and shade he often got confused about who the face belonged to.

If S were the way the fictional Funes was described as being, this
should not have been true of faces, but of any and all kinds of things:
They should not be able to abstract from the concrete sensorimotor
instants any kind of pattern or regularity. Hence they should be unable
to learn, or in any way behave differentially depending on anything that
had happened.

> 'Funes the memorious'. Funes describes his
> memory as a garbage heap. He said that:

If Funes's memory were really infinite, he couldn't even say that,
because it depends on abstracting what a "memory" is and what a "garbage
heap" is from the endless garbage heap of videos...

> he replaced
> numbers such as 7014 with things such as 'the rail road'. He couldn't
> comprehend that these expansive incoherent terms were the precise
> opposite to a system of numbers.

If Funes's memory were really infinite, all he could do would be to give
one unique name to each unique instant. No "7014" or "railroad" except
as a proper name, such as Donna or Nick...

> It has to be said therefore that Funes
> was almost incapable of ideas of a general platonic sort.

He should have been not almost, but totally incapable of any abstraction
at all; not just incapable to the more abstract abstractions, such as
numbers. Even naming colors or shapes (or people) requires abstracting
the relevant features they all share and ignoring the irrelevant ones they
don't share.

> This of
> course caused problems in his everyday understanding of things. He also
> thought it was nonsense for a dog to have the same name no matter what
> position it was in. He thought that the dog's name should change
> according to how it was sitting and what direction it was seen from.

And he shouldn't even know "it" was a "dog."



This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:24:15 GMT