Working memory

Sturla Molden stumol at
Fri Mar 20 15:45:12 EST 1998

On 20 Mar 1998 03:49:22 GMT, flefever at Frank LeFever)

>LTP has been suggested chiefly as the basis for formation of long-term
>memories, not "working memory" in the sense I have emphasized
>(initiating a process which continues over a rather long period).

LTP has been suggested to be involved in memories that rely on
neural plasticity rather than dynamic change in neuronal activity.
This also include some memories that are task-specific and not
permanently retained. Such memories require continously updating. This
was what Olton (1985) ment by his working memory view of hippocampal
function. And it is and operational definition. E.g. you go to the
store and remember to by a bottle of milk, but two days later it's all
gone. If Patient HM was to purchase a bottle of milk, he would have
forgot why he went to the store when he got there. 

I think in these sence, "working memory" is more like an "intermediate
term memory", enduring but futile. Whether such working-memory has a
physiological foundation that differs from long-term memories (that
are not lost, ore are they?) is also an unsettled question.
Neccesarily, long-term memories are shuttled through a hypothetical
intermediate-term store (Rawlins' [1985] buffer?), so interfering with
working-memory also interfer with  long-term memory. There are also an
interpretational problem with this categorical model of memory. The
"power-law of memory" from cognitive psychology doesn't yield this
prediction. My guess is that working-memories in this sence is just
long-term memories that are lost with a higher rate than the memories
we term proper "long-term memories". In this sence there are no
completely permanent memories, memories are just lost at a different
rate (some very slow). Speaking in favour of the categorical model,
electroconvulsive therapy also cause some retrograde amnesia, hower,
this may also be interpretated  in the way that these memories are
just weaker (the reason why they are lost at a faster rate) and
therefore more affected by the ECT. An interesting question is what
modulate(s) the persistance of long-term memories.

The distinction between short-term memories and long-term memories are
more clear-cut. Short-term memories are dependant on dynamic changes
in neuronal activity, long-term memories require plasticity. But from
my point of view, splitting up enduring (plasticity-dependant)
memories into different classes according to their stored time is a
conceptual construct that may prove to be devoid of a true
physiological foundation.

And speaking of memeories, are there any operational definitions of
"impicite/procedural" and "explicite/declarative" memories? I think
the common aggreement that "memories we are conscious about are the
explicite ones" is a bit sloppy, afterall, what does it mean -
physiologically speaking - to be mentally "conscious" of something? I
would like to see a definiton that  relate the two forms of memories
to something that can be objectively measured (in both man and
animals). I would be grateful if anyone would give me a reference to
litterature dealing with this  matter in a proper way. 

Sturla Molden

More information about the Neur-sci mailing list