My understanding of Baddeley's model is that it continues to maintain the
structural distinction between short-term and long-term memory proposed by
the Atkinson/Shiffrin "modal model" (sensory registers, STM, and LTM).
Baddeley has taken STM and created a working memory system comprised of a
Central Executive and various "slave systems" - the visual-spacial sketch
pad and the rehearsal loop system (itself composed of two systems: the
articulatory loop and the phonological store) being the two that have been
explored experimentally. Thus, the "traditional" idea of the STM structure
is now instantiated as a complex of separate processing structures. I
don't think it is useful to ask which component of Baddeley's model
corresponds to the old STM structure. Performance on a given immediate
memory task (which used to reflect the operation of STM) depends on the
extent to which individual components of the working memory system can be
used to maintain information.
For example, digit span performance reflects storage of digits in the
Central Executive as well as in the rehearsal loop. The contribution to
digit span performance made by all or part of the rehearsal loop can be
eliminated by various experimental manipulations. For example, forcing
subjects to articulate irrrelevant material (e.g., known as articulatory
suppression, in which the subject must repeat something like "dah, dah,
dah" while engaged in the memory task) will elimnate both the articulatory
loop (because it is assumed to be based in motor programs used for
producing speech) and the phonological store (because the sound of your own
voice speaking fills the phonological store). The result is a reduction in
digit span, loss of word-length effects on memory span, and elimination of
acoustic confusion errors in performance. Forcing subjects to listen to
irrelevant speech, however, eliminates the phonological store without
interferering with operation of the articulatory loop. This manipulation
will eliminate acoutical confusion errors & reduce the digit span (but not
as much as is obtained with articulatory suppression, because the subject
can still use the articulatory loop for storage), but not eliminate the
word-length effect (because this is based on articulation time).
The possibility of additional "slave" systems is left open. Reisberg &
others are exploring the possibilities of another system that can code
verbal information in terms of hand movements. The emergence of new
storage sub-systems with specialized training raises questions about the
_structural_ status of these systems. Reisberg discusses research in
which subjects can increase their digit span when they learn a system in
which digits are coded in terms of finger-tapping. The system can be
disabled by engaging in the equivalent of articulatory suppression -
drumming the fingers. There is speculation that training in ASL might also
establish motor programs for hand movements as a system that is analogous
to (but separate from) the articulatory loop.
(This is what you get for asking a cognitive psychologist to explain this -
we get carried away!)
Claudia
Reisberg, D., Rappaport, I., & O'Shaughnessy, M. (1984). Limits of
working memory: The digit digit-span. Journal of Experimental Psychology:
Learning, Memory, & Cognition, 10, 203-221 .
___________________________________________________
Claudia J. Stanny, Ph.D. e-mail: [EMAIL PROTECTED]
Department of Psychology Phone: (850) 474 - 3163
University of West Florida FAX: (850) 857 - 6060
Pensacola, FL 32514 - 5751
Web: http://www.uwf.edu/~psych/stanny.html
