Working memory: good old STM?

F. Frank LeFever flefever at
Tue Mar 24 23:35:36 EST 1998

Ah, Gutnikov!  of the NMDA and LTP work!

Anyone may define anny term one wants, but as defined by the clinical
phenomena and the neuropsychogical experimental operations I have
focused on, the time constraint does seem to be important--so
remembering location of your car as it changes from one day to the next
would not exemplify working memory as defined by them.  However,
remembering the positions of other cars, as they change, while you
drive among them would.

I have, however, often thought of the importance of the longer-term 
updating you describe, given that I must find a new parking place every
day, on the streets of New York City, and have sometimes considered
studying this formally.

There is definitely a point of coordination between your definition and
the other I have emphasized, however.  Although usually not studied  in
exactly those terms (updating car locations, etc. from day to day),
there is reason to think that this too must depend on prefrontal
integrity, as an example of proper location of memory items in time. 
Failure in this is thought to be one basis for "confabulation"--i.e., a
patient may give an answer which would be correct in another time or
place but not the time and place of the question.

I have done some research (presented at TENNET and at INS meetings,
partially published in _Brain & Cognition_) on temporal ordering or
temporal locating of memory, using a test I have devised (the "Recency
Test"); and Thomas Benke (Innsbruch) has presented patient data  using
this and other tests at a meeting of a European neuroscience society.

I am embarassed to ask, because I have never understood the
significance of LTP apparently being dominant in the cerebrum and LTD 
in the cerebellum (if indeed this is correct), but--might LTD (i.e.
long-term depression) be involved in the inhibition necessary for

Frank LeFever

In <6f8jgj$jqk$2 at> SG <"SG"@$$$$$$> writes:

>F. Frank LeFever wrote:
>> (2) The emphasis is not on memory per se but on USE of memory,
>> involving momentary retrieval from long-term stores as well as
>> short-term or immediate.  This includes timely "forgetting" of one
>> so that it can be replaced by another at the proper time, and thus
>> ongoing selective enhancement of one vs. another competing memory OR
>> against the competition of an EXTERNAL stimulus (cf. the concept of
>> "stimulus bound" behavior).
>Working memory is first of all *associative* memory. 
>Associative memory means that associations between some items is
>involved. I use the word "item" in a broad sense: it can be some
>or complex sensory input, or internal representation of sensory inputs
>(such as memory for a particular environment or episode), a concept,
>The other type of associative memory is "reference memory".
>Working memory critically differs from reference memory in the
>REFERENCE memory: Association between items remain the same throughout
>time. Every time the items (and their associations) are recalled, the
>association becomes stronger.
>WORKING memory: Associations change or reverse, thereby making old
>associations useless or even contradictory to the current true state
>affairs. Repetitive recall of the items and their associations does
>help to build a stronger association. On the contrary, old
>must be effectively inhibited.
>----- An example: 
>The location of your nearest supermarket and its car park is stored in
>your reference memory. The more often you go there the better you know
>the route. 
>The car park is crowded and you have to park your car in a different
>place every time you go shopping. The location of your car in the car
>park is kept in your working memory. Every time the items are the same
>object "car" and environment "car park", but remembering where your
>was parked last time doesn't help you to find the car today. Quite the
>opposite, for successful return to the car you must suppress, or
>inhibit, the old association.
>----- end of example
>So, the critical feature of working memory is inhibition of old
>associations rather than the time scale. This makes the terms "working
>memory" and "short-term memory" not interchangeable (though often
>working memory operates on the same time scale with short-term
>Thus, my definition of working memory is:
>WORKING MEMORY is acquiring, storage and recall of new associations
>are to replace or to inhibit earlier, now irrelevant or contradictory,
>associations related to the same item(s).
>----- Proposal:
>As the term "working memory" is currently used in many senses, shall
>use this opportunity and collectively try to compile a list of
>If there will be support to this idea, I will dig some proper
>references from my old notes.
>Best regards,
>Sergei Gutnikov
>remove $$$$$$ from the e-mail address

More information about the Neur-sci mailing list