Working Memory

Working memory is the cognitive system that allows us to keep active a limited amount of information (roughly, 7 ± 2 items) for a brief period of time (roughly, a few seconds). This system has been a major research topic since the advent of the cognitive revolution in the 1950s, and was earlier referred to as "short-term memory." It was then thought to have two functions: storing material that we have to recall in a few seconds, as when we rehearse a phone number until we dial it, and providing a gateway to long-term memory (e.g., Atkinson and Shiffrin 1968). While cognitive scientists continue to believe in the simple storage purpose, their belief in the gateway function has been somewhat undermined by the existence of neurological patients who are impaired in short-term memory tasks, but perform normally on long-term memory tasks (see, e.g., Shallice 1988). Rather, cognitive scientists now assume that the major function of the system in question is to temporarily store the outcomes of intermediate computations when PROBLEM SOLVING, and to perform further computations on these temporary outcomes (e.g., Baddeley 1986). For example, when mentally multiplying two-digit numbers like 38 × 19, we may first compute and store the partial product 8 × 9 = 72, later use this partial product in further computations, and subsequently drop it when it is no longer needed. Given this role, the system in question has been renamed "working memory," and is considered critical not only for analyzing MEMORY, but for understanding thought itself.

In what follows, first we review some basic characteristics of working memory, and then consider its role in higher-level cognition. We mention empirical evidence from various human studies, including cognitive-behavioral experiments, neuropsychological (patient) studies, and neuroimaging experiments (using POSITRON EMISSION TOMOGRAPHY or functional MAGNETIC RESONANCE IMAGING).

There appear to be different working memories for different kinds of materials, particularly different systems for verbal and spatial information. In a paradigm cognitive-behavioral experiment, subjects perform a working-memory task while concurrently performing a secondary task. Any secondary task usually causes some interference with a working-memory task, but verbal secondary tasks interfere more with verbal than with spatial working-memory tasks, whereas spatial secondary tasks interfere more with spatial than with verbal working-memory tasks (e.g., Brooks 1968). This pattern of selective interference supports the hypothesis of separate systems for verbal and spatial working memory. (These separate working-memory systems may be connected to separate verbal and spatial perceptual systems.)

The above results are bolstered by neuropsychological findings. There are pairs of neurological patients such that one is impaired on a standard measure of verbal working memory -- digit span -- but normal on a standard test of spatial working memory -- Corsi blocks test -- whereas the other patient shows the reverse pattern (see McCarthy and Warrington 1990). This double-dissociation between verbal and spatial working-memory tasks argues for two separate systems. Perhaps the most direct evidence for two systems comes from neuroimaging experiments. Subjects perform either a verbal recognition test -- for example, remembering the names of four letters for 3 sec -- or a spatial recognition test -- for example, remembering the locations of three dots for 3 sec -- while having their brains scanned. Different areas of the brain are activated in the two tasks, with almost all of the activations in the verbal task being in the left hemisphere, and most of the activations in the spatial task being in the right hemisphere (Smith, Jonides, and Koeppe 1996). (Other neuroimaging studies indicate that there might be separate working memories for spatial and visual-object information, just as the single-cell evidence shows for nonhuman primates -- see WORKING MEMORY, NEURAL BASIS OF).

Within verbal and spatial working memory, there is evidence for a further subdivision, that between a passive storage process and an active rehearsal process. The evidence is strongest for the verbal system. In cognitive-behavioral studies, experimenters have argued that some effects reflect only a storage process, for example the phonological similarity effect, in which the short-term recall of words is poorer for phonologically similar than phonologically dissimilar ones (Conrad 1970) whereas other effects are due to rehearsal, for example, the word-length effect, in which the short-term recall of words declines with the time it takes to say the words (Baddeley, Thompson, and Buchanan 1975). Importantly, when subjects doing these tasks are prevented from rehearsing the words by having to articulate some irrelevant word or phrase, the word-length effect disappears but the phonological-similarity effect remains intact (Longoni, Richardson, and Aiello 1993). Presumably, the irrelevant articulation blocked rehearsal, but had no effect on the storage buffer. Further support for this interpretation comes from the study of a patient whose brain damage presumably disrupted only the rehearsal component. This patient shows a normal phonological-similarity effect, but no effect of word length or of irrelevant articulation (Basso et al. 1982).

There is converging evidence for the storage-rehearsal distinction from neuroimaging studies. Subjects are scanned while doing a short-term recognition task, which presumably involves storage plus rehearsal, or while doing a task that involves only articulation or rehearsal. Both tasks activate areas in the left-hemisphere frontal cortex that are known to be involved in the planning of speech, whereas only the memory task activates posterior-parietal regions thought to be involved in storage per se (Paulesu, Frith, and Frackowiak 1993; Awh et al. 1996). Recent neuroimaging experiments argue for a comparable storage-rehearsal distinction in spatial working memory, where spatial rehearsal appears to amount to selectively attending to particular locations (Awh and Jonides forthcoming).

Some of the best evidence for (verbal) working memory playing a role in higher-level cognition comes from cognitive- behavioral studies. One line of evidence is that there are substantial correlations between (1) a measure of a person's verbal working-memory capacity -- the reading-span task -- and (2) the person's performance on either a reasoning task -- the Raven Progressive Matrices test -- or language-understanding tasks (e.g., Carpenter, Just, and Shell 1990; Just and Carpenter 1992). A second piece of behavioral evidence is the finding that performing a working-memory task interferes with a concurrent reasoning task (solving syllogisms) more than does performing a non - working-memory task (Gilhooly et al. 1993). Again, there is converging evidence from recent neuroimaging experiments. When people engage in either a reasoning task (the Raven test again) or a complex-categorization task, many of the areas found active are those activated in standard working-memory studies (Prabhakaran et al. 1997; Smith, Patalano, and Jonides 1998).

Other persuasive evidence for working memory's role in higher-level cognition comes from computational research, specifically the use of symbolic models to simulate higher-cognitive processes. Simulations of this sort routinely give a major role to working-memory operations, and provide a detailed account of exactly how working memory can be used to regulate the flow of information processing during CATEGORIZATION, PLANNING, reasoning, PROBLEM SOLVING, and language understanding (e.g., Anderson 1983; Newell 1990; Carpenter et al. 1990).

Additional links

-- Edward E. Smith

References

Anderson, J. R. (1983). The Architecture of Cognition. Cambridge, MA: Harvard University Press.

Atkinson, R. C., and R. M. Shiffrin. (1968). Human memory: A proposed system and its control processes. In K. W. Spence, and J. T. Spence, Eds., The Psychology of Learning and Motivation: Advances in Research and Theory, vol. 2. New York: Academic Press.

Awh, E., and J. Jonides. (1998). Spatial selective attention and spatial working memory. In R. Parasuraman, Ed., The Attentive Brain. Cambridge, MA: MIT Press, pp. 353-380.

Awh, E., J. Jonides, E. E. Smith, E. H. Schumacher, R. A. Koeppe, and S. Katz. (1996). Dissociation of storage and rehearsal in verbal working memory: Evidence from PET. Psychological Science 7:25-31.

Baddeley, A. D. (1986). Working Memory. Oxford: Oxford University Press.

Baddeley, A. D., and G. J. Hitch. (1974). Working memory. In G. Bower, Ed., Recent Advances in Learning and Motivation, vol. 8. New York: Academic Press.

Baddeley, A. D., N. Thompson, and M. Buchanan. (1975). Word length and the structure of short-term memory. Journal of Verbal Learning and Verbal Behavior 14:575-589.

Basso, A. H., H. Spinnler, G. Vallar, and E. Zanobia. (1982). Left hemisphere damage and selective impairment of auditory verbal short-term memory: A case study. Neuropsychologica 20:263-274.

Brooks, L. R. (1968). Spatial and verbal components of the act of recall. Canadian Journal of Psychology 22:349-368.

Carpenter, P. A., M. A. Just, and P. Shell. (1990). What one intelligence test measures: A theoretical account of the processing in the Raven Progressive Matrices Test. Psychological Review 97:404-431.

Conrad, R. (1970). Short-term memory processes in the deaf. British Journal of Psychology 61:179-195.

Gilhooly, K. J., R. H. Logie, N. E. Wetherick, and V. Wynn. (1993). Working memory and strategies in syllogistic- reasoning tasks. Memory and Cognition 21:115-124.

Just, M. A., and P. A. Carpenter. (1992). A capacity theory of comprehension: Individual differences in working memory. Psychological Review 99:122-149.

McCarthy, R. A., and E. K. Warrington. (1990). Cognitive Neuro-psychology: A Clinical Introduction. San Diego: Academic Press.

Newell, A. (1990). Unified Theories of Cognition. Cambridge, MA: Harvard University Press.

Paulesu, E., C. D. Frith, and R. S. J. Frackowiak. (1993). The neural correlates of the verbal component of working memory. Nature 362:342-344.

Prabhakaran, V., J. A. L. Smith, J. E. Desmond, G. H. Glover, and J. D. E. Gabrieli. (1997). Neural substrates of fluid reasoning: An fMRI study of neocortical activation during performance of the Ravens Progressive Matrices Test. Cognitive Psychology 33:43-63.

Shallice, T. (1988). From Neuropsychology to Mental Structure. Cambridge: Cambridge University Press.

Smith, E. E., J. Jonides, and R. A. Koeppe. (1996). Dissociating verbal and spatial working memory using PET. Cerebral Cortex 6:11-20.

Smith, E. E., A. Patalano, and J. Jonides. (1998). Alternative strategies of categorization. Cognition 65:167-196.

Working Memory

See also

Additional links

References

Further Readings