1.1.1 Experimental Obs.
point to 2 STM + LTM
coding info in different
188.8.131.52 STM Info held in accoustic or
speech based form. LTM
coded in terms of meaning
184.108.40.206.1 Brown (1958)
rates of briefly
compared to better
supporting idea that
STM is more labile than LTM
220.127.116.11.2 Span of Immediate
Memory is limited
to a few items,
18.104.22.168.3 M. Span is longest
sequence that can be
22.214.171.124.3.1 Low limit on span
from LTM on
grounds of Ltd
126.96.36.199.4 This evidence at time V.
compelling. Led to the two
store models, common
features referred to as
'Modal Model' (Murdock,
188.8.131.52.4.1 Main Assumptions of 'MM':
1:STM limited capacity store of
short duration. 2: control
processes like subvocal
rehearsal can be used to keep
info in STM. 3: Info in STM
gradually transferred to LTM.
See Atkinson & Shifrin 1971 Modal Model of memory, attached.
184.108.40.206.4.1.1 While Model Model
authors argued that
provided by STM
crucial for cognitive
Cognitive activities like sentence comprehension or problem solving.
220.127.116.11.18.104.22.168 In other words, STM behaves
as some sort of working
Can gain insight into plausibility of this through trying to understand garden path sentences. Just and Carpenter (1992) argued one keeps several interpretations of sentence in working memory. Caplan and Waters (1999) proposed that comprehension draws on more specialised resources than working memory.
Modal model captures some important insights, but consensus was fleeting. One concern was whether various strands of evidence for distinguishing STM and LTM converged on a coherent account---eg. different ways of estimating capacity of STM gave different answers, and the reasons for this were unclear. Consequently a resurgence in interest in LTM (eg 'levels of processing approach' by Craik and Lockhart 1972) occurred. Another concern was whether STM does actually work as a working memory, Neuropsychological evidence from a patient KF (brain damaged) supported some interpretations of modal model but posed difficulties for it in other ways. pg 156 MTB.
22.214.171.124.126.96.36.199.1 WORKING MEMORY AS MORE THAN STM?
188.8.131.52.184.108.40.206.1.1 Due to KF difficulties,
Baddeley and Hitch
(1974) made empirical
whether STM acts as
220.127.116.11.18.104.22.168.1.1.1 They used Dual Task Paradigm
People performed 2 tasks at same time. Logic is that 2 tasks will interfere with one another if they need access to a common resource and if their combined demands exceed its capacity. pg 156-157 MTB
22.214.171.124.126.96.36.199.188.8.131.52 B & H (1974) drew 2
conclusions: 1: Common
system combines temp. info
storage with ongoing mental
operations. 2: Working
memory includes an
additional resource that is not
shared with STM
1: The finding that an irrelevant STM task interferes with a range of cognitive tasks (supporting conc. 1)
2: When load on STM approached memory span and therefor filled short term storage capacity, there was no catastrophic breakdown in concurrent cognition (supporting conc. 2). pg 157 MTB
184.108.40.206.220.127.116.11.1.2 Further evidence for a
distinction between STM
and working mem. came
from studies of individual
Logic is that if two tasks involve similar psychological processes, a person who performs well at one should do well on another. In statistical terms the two abilities should be positively correlated.
18.104.22.168.22.214.171.124.1.2.1 Daneman & Carpenter (1980) argued that
standard measures of STM like word span
tax storage capacity but don't assess
capacity to combine storage with ongoing
126.96.36.199.188.8.131.52.184.108.40.206 Daneman & Carpenter devised reading span task. They interpreted
their findings as showing that WM Capacity is an important source
of individual differences in language comprehension, the key
characteristic of WM being combining temporary storage with
information processing. In line with Baddeley and Hitch 1974.
220.127.116.11.18.104.22.168.22.214.171.124.1 A criticism of D & C approach was that their correlations might be an artifact
of similarities in processing operations in the various tasks they use ( reading
span/listening span & language comprehensionall involve language
comprehension whereas wordspan does not.
126.96.36.199.188.8.131.52.184.108.40.206.1.1 So Turner & Engle (1989) devised an operation span task in
which participants solved sets of arithmetical calculations.
They found Operation span a superior predictor of reading
comprehension than standard STM span inspite of involving
dissimilar processing operations.
220.127.116.11.18.104.22.168.22.214.171.124.1.1.1 Their results therefore support
idea of general working memory
system that is common to a range
of different activities involving the
combination of info processing with
temporary storage. Subsequent
work by Engle et al (1999b) has
shown that WM span more closely
related to general intelligence than
126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11 NEW MAP: STRUCTURE OF WORKING MEMORY