> On Mon, 24 Feb 1997, zaimoni@ksu.edu wrote:
>
> >Do *not* assume that #1 *knows* what really happened after she starts
> >discussing it with the other two! The evidence suggests that she
> >completely forgets what "really happened", and becomes just as clueless
> >as the other two.
>
> OK, I phrased that badly. I'd kinda figured that was what was happening.
> My real question is; how the hell does this happen in so short a period
> of time? or; why are poeples memories so damned unreliable?
[Finally, time loading decompresses below hypercritical....]
Assume that memory is stored physically in the human brain, for a
moment. [It definitely *is* accessed that way, so perhaps this reasoning
generalizes to the blatantly measurable situation.]
What is the reliability of a memory system that stores multiple "data
items" in the *same* physical hardware?
Should this work at all???
The last is a loaded question. It turns out that not only does it work,
it can be made to work in electronics [neural network hardware].
So, if the memory of the conversation results is sufficiently reinforced,
it can literally overwrite and annihilate the memory of what actually
occurred. In fact, this is how software neural networks are instructed.
[Fuzzily, anyway. Anything that requires a PhD thesis to document isn't
easy to describe.]
//////////////////////////////////////////////////////////////////////////
/ Towards the conversion of data into information....
/
/ Kenneth Boyd
//////////////////////////////////////////////////////////////////////////