Development of Working Memory: Should the Pascual- Leone and the Baddeley and Hitch Models Be Merged?

Journal of Experimental Child Psychology 77, 128 137 (2000) doi:10.1006/jecp.2000.2592, available online at http://www.idealibrary.com on Development of Working Memory: Should the Pascual- Leone and the Baddeley and Hitch Models Be Merged? Alan D. Baddeley University of Bristol, Bristol, United Kingdom and Graham J. Hitch Lancaster University, Lancaster, United Kingdom The data presented by Kemps, De Rammelaere, and Desmet (2000, this issue) appear to have some aspects that fit most readily into our own model (Baddeley & Hitch, 1974), while others appear to support that of Pascual-Leone (1970). We accept that our initial model said little about development and was better able to account for relatively simple memory-based tasks than more complex cognitive activities. More recent elaborations of the model are, however, able to throw new light on the processes underlying cognitive development, offering a better account than that provided by existing neo-piagetian interpretations. Meanwhile, the addition of a fourth component to the model, namely the episodic buffer, offers a way of dealing with more complex cognitive activities. Given the major differences between our own model and that of Pascual-Leone in basic assumptions, and in theoretical style, we suggest that any attempt to combine the two would be premature. 2000 Academic Press Key Words: children s memory; working memory. Kemps, De Rammelaere, and Desmet (2000, this issue) describe a study of working memory in children which has the aim of demonstrating the complementarity of the working memory models that have been postulated by Pascual- Leone and Baddeley and Hitch. Using tasks in which the child must remember where to place an array of colored dots on a cartoon figure of a clown (Mr. Peanut), or movements to a sequence of spatial locations (Corsi blocks), they conclude that effects of verbal and visuo-spatial interference are best accounted for by the Baddeley and Hitch model, while Pascual-Leone s neo-piagetian model gives a better account of the pattern of developmental changes. This Address correspondence and reprint requests to either Alan Baddeley, Department of Psychology, University of Bristol, BS8 ITN, United Kingdom, or Graham Hitch, now at the Department of Psychology, University of York, YO10 5DD, United Kingdom. 0022-0965/00 $35.00 Copyright 2000 by Academic Press All rights of reproduction in any form reserved. 128

REFLECTIONS ON KEMPS ET AL. 129 reflects the greater emphasis of the Baddeley and Hitch model on the subsidiary visual and verbal slave systems of working memory, rather than the central executive, in contrast to Pascual-Leone s emphasis on more complex and executive aspects of cognition. Should the two models therefore be combined so as to take advantage of the best features of each? Let us begin by accepting that the central executive is the least well understood component of the Baddeley and Hitch model. Furthermore, the model was developed to account for adult data; while there is now considerable research on its application to the processes of development, the need to account for developmental change is less central to the model than to that of Pascual-Leone. There are, however, major differences between the two approaches in theoretical style, suggesting that an attempt to combine them would be worthwhile only if the potential advantages were clear. We begin by outlining these differences, and then we go on to discuss the strength of the evidence presented by Kemps et al. before proceeding to discuss the way in which the Baddeley and Hitch model might be adapted to account for the evidence from this and other developmental studies. The Baddeley and Hitch model represents an elaboration of earlier unitary models of short-term memory (STM) proposed by Broadbent (1958) and Atkinson and Shiffrin (1968). It attempts to account for a wide range of data in terms of a few broad concepts, elaborating those concepts only when it is clear that simpler accounts are not viable. The model currently has three major components. One is a central control system, the central executive. It is aided by two subsidiary systems, the phonological loop and the visuo-spatial sketchpad. As we shall see below a fourth component, the episodic buffer, has recently been proposed (Baddeley, in press). While a number of precise and well-specified models of the phonological loop have been developed (Burgess & Hitch, 1992, 1999; Henson, 1998; Page & Norris, 1998), the overall framework remains relatively simple. In contrast, Pascual-Leone s model has multiple components including numerous general-purpose and specific schemes that vary in activation, and the operation of the model is developed mathematically. The comparative simplicity of the Baddeley and Hitch model has led to its application within neuropsychology, and to its investigation using neuroradiological imaging techniques (see Smith & Jonides, 1998, 1999, for reviews). Neurological evidence has in turn fed into a gradual process of evolution, as illustrated, for example, by the proposal for the new episodic buffer component (Baddeley, in press). Given the complexity of the Pascual-Leone model, with the operation of many of the components still apparently poorly specified, it is less clear how it could be applied neuropsychologically. Its link to Piaget s theory of developmental stages may also be seen as a drawback by some investigators, given the substantial controversy that still surrounds this approach to development (see, e.g., Chandler & Chapman, 1991). The evidence produced by Kemps et al. does seem to fit neatly into the

130 BADDELEY AND HITCH Baddeley and Hitch model. In particular, the interaction between the effects of articulatory suppression and age is consistent with extensive evidence that strategic use of the phonological loop typically occurs only after the age of 8 or 9 years (Hitch, Halliday, Dodd, & Littler, 1989; Hitch, Woodin, & Baker, 1989). The effects of articulatory and spatial suppression on the Corsi blocks task are also consistent with the assumption of separate slave systems. These data are inconsistent with the proposal of a single common memory system in which articulatory suppression has a general interfering effect; see for example Macken and Jones (1995) and Neath and Nairne (1995). How strong is the evidence presented by Kemp et al. for the Pascual-Leone model? Their claim for support rests principally on the fact that performance develops with age, and that it deviates from a linear function in a prespecified way. More specifically, the Pascual-Leone model predicts that performance rises through alternating steps and plateaus spanning successive 2-year intervals. A good deal of Kemp et al. s data, however, could be accounted for by simply noting that groups that are separated by 2 years in age are more likely to give significantly different performance than those separated by 1. This admittedly does not give a complete account of the data, but neither does the Pascual-Leone theory. What would be needed to demonstrate convincingly that the shape of the function was exactly as predicted? It would not be sufficient to demonstrate that the developmental difference between one pair of ages was significant and another was not: This could be produced simply by greater variance at one particular age. What is needed is to demonstrate a significant difference in slope between stages over which development is assumed to be stepwise, compared with the slope across those ages over which a performance plateau is predicted. No doubt other, more convincing evidence may become available, but at the present we do not feel that it is yet strong enough to encourage us to incorporate the complexities of the Pascual-Leone approach in our own model. We note with interest a recent version of Pascual-Leone s model that incorporates ideas about the phonological loop (Morra, 2000). The model is indeed very complex, and despite this it does not address visuo-spatial working memory or neuropsychological data. So what should we do instead? Our own research strategy has been to continue to explore developmental applications of the Baddeley and Hitch model. This has proved particularly fruitful in demonstrating the developmental fractionation of function whereby changes in the phonological loop and the visuo-spatial sketchpad have been shown to follow different developmental pathways (Hitch, 1990). Thus, as Kemps et al. s data confirm, the phonological loop is used more pervasively as development proceeds, with an increasing tendency to recode nonverbal stimuli. The strength of the bias toward verbal recoding is demonstrated by the observation that adults ability to form a mental image of a nameable picture is improved when subvocal articulation is suppressed (see, e.g., Brandimonte, Hitch, & Bishop, 1992). The fractionation of the phonological loop and the

REFLECTIONS ON KEMPS ET AL. 131 visuo-spatial sketchpad has also proved useful in analyzing developmental memory deficits associated with different types of genetic disorders. Thus, children with Down syndrome are impaired on tasks that involve the phonological loop but are relatively spared on tasks involving the visuo-spatial sketchpad, whereas children with Williams syndrome show the converse pattern (Jarrold, Baddeley, & Hewes, 1999; Wang & Bellugi, 1994). Such a dissociation of functional weaknesses in working memory may underpin the different patterns of cognitive deficit associated with Down and Williams syndromes. A consequence of using the Baddeley and Hitch model to study development is that this in turn provides further data on the validity of the model. A clear example is provided by the word length effect, the tendency for sequences of words with longer durations to be less well remembered in immediate serial recall tasks (Baddeley, Thomson, & Buchanan, 1975). In the original Baddeley and Hitch model, this finding was explained by assuming that decay of memory traces in the phonological loop is offset by subvocal rehearsal, a process that takes longer for longer words. Initial developmental data fitted such an account, as children s ability to recall lists of long and short words at different ages was found to be linearly related to the speeds at which they could rehearse the items (Nicolson, 1981). However, further developmental studies showed that children as young as 4 are sensitive to word length when recalling sequences of spoken words, in much the same way as older children and adults (Hulme, Thomson, Muir, & Lawrence, 1984). As this age is somewhat before children acquire the ability to rehearse (Gathercole, Adams, & Hitch, 1994), 4-year-olds sensitivity to word length questions the explanation given by the Baddeley and Hitch model. Subsequent work showed that, unlike in serial recall, young children do not show a word length effect in probed recall, where the response to the probe is a single item (Henry, 1991). This observation suggests that the word length effect is partly due to the output delays that occur during recall of an entire sequence. Cowan et al. (1992) confirmed that different output delays for long and short words could affect recall by studying memory for lists containing mixtures of long and short items. Furthermore, simulations of the phonological loop using connectionist modeling have successfully combined the effects of output delays and rehearsal in a mechanistic account of the word length effect (Burgess & Hitch, 1992, 1999; see also Brown & Hulme, 1995). In the case of the word length effect, therefore, the Baddeley and Hitch model has not only helped describe developmental differences but has in turn benefited from the attempt to account for children s behavior. In other work we have explored the development of the ability to combine temporary information storage with ongoing processing. We have undertaken this in the context of a variant of the Pascual-Leone model proposed by Case and colleagues (Case, Kurland, & Goldberg, 1982). Case et al. devised a counting span task similar to the well-known reading span and listening span tasks developed by Daneman and Carpenter (1980). These tasks are different ways of

132 BADDELEY AND HITCH assessing working memory span, a measure of how much temporary information can be maintained while performing processing operations. Case et al. found that counting span increased during development, and that the amount of growth was predicted by the increase in the speed of processing with age. At the center of the theoretical account given by Case et al. is the idea that working memory span reflects the capacity to share resources between the dual functions of processing and storage. Processing is assumed to become faster and more automatic as development proceeds, freeing more resources for storage and hence increasing working memory span. However, Case et al. did not directly observe the resource demands of processing, but rather inferred them from changes in processing speed. A simple and plausible alternative account is that faster processing results in higher working memory span because temporary information has to be maintained for less time (Hitch & Towse, 1995). According to this alternative interpretation, working memory span is limited by the rate at which recently stored information loses activation. Recent experimental evidence supports such a reinterpretation. Thus, children s working memory spans are reduced when retention durations are increased while the resource demands of processing are held constant (Hitch, Towse, & Hutton, in press; Towse, Hitch, & Hutton, 1998). Conversely, working memory span in children is insensitive to the attentional demands of processing operations when time taken to complete processing is held constant (Towse & Hitch, 1995). We note that these data present difficulties not only for Case et al. s model, but also for the original Baddeley and Hitch model, in which the executive was assumed to share resources between processing operations and temporary information storage in a similar way. Thus, as for the word length effect, data from studying children have challenged part of the original Baddeley and Hitch model. We note, however, that the concept of the executive has undergone substantial revision (Baddeley, 1986, 1996) in order to take into account evidence from a variety of other sources, as will be described later. A further reason for tackling the question of development stemmed from our attempts to explore the functional significance of the phonological loop. Adult patients with a pure but substantial disruption to the functioning of the loop seem to have remarkably few problems in general cognitive functioning, raising the question of whether the loop is of more than minor significance. When they are studied in more detail, however, it becomes clear that patients with very specific loop deficits have considerable difficulty in new phonological learning such as would be required for learning the vocabulary of an unfamiliar language (Baddeley, Papagno, & Vallar, 1988). This finding suggests that the system may have evolved for the acquisition of language, a view that is further supported by work on second language learning in normal adults, and on the acquisition of native language vocabulary by both normal and specific language impaired children (see Baddeley, Gathercole, & Papagno, 1998, for a review). These results raise the question of the underlying mechanism. Snowling and

REFLECTIONS ON KEMPS ET AL. 133 her collaborators, for example, suggest that the developmental data may reflect some form of general phonological processing deficit, rather than a phonological loop limitation (Snowling, Chiat, & Hulme, 1991). This account does, however, seem to have difficulty in explaining the extensive supportive data from adults who have acquired a phonological loop deficit when mature, or whose phonological loop operation is disrupted by articulatory suppression (Baddeley et al., 1998). Gathercole (1995) and her colleagues have thrown further light on the mechanisms underlying the operation of the loop by studying the effects of wordlikeness on the immediate recall of pseudowords. Wordlikeness is simply the degree to which a pseudoword resembles known words. Gathercole found that the more wordlike an item is, the greater the probability that it will be correctly recalled, suggesting a mechanism based on the activation of long-term memory (LTM). However, the subsequent acquisition of vocabulary by a child is better predicted by performance on the less wordlike items (Gathercole, 1995). Furthermore, whereas serial recall is strongly influenced by how wordlike the items are (Gathercole, Frankish, Pickering, & Peaker, 1999), serial order recognition is comparatively unaffected by lexicality, suggesting an effect that occurs at output rather than storage (Gathercole, Pickering, Hall, & Peaker, 2000). Finally, performance on this recognition version of the task, despite its comparative independence from long-term learning, nevertheless correlates highly with vocabulary scores in both English (Gathercole, Service, Hitch, Adams, & Martin, 1999) and Catalan (Baques, personal communication, 2000). This would seem to point to some form of storage that can feed long-term phonological memory, but is not dependent on it, except for the output processes involved in recall. We hope these examples will help demonstrate that, although the Baddeley and Hitch model was not designed to study development, attempting to give an account of developmental change within the model continues to be highly fruitful. We suggest that the robustness and simplicity of the model s concepts and methods have contributed to the insights it has generated, and that these provide a solid empirical and theoretical base for extending our understanding of developmental change. However, as noted earlier, a major advantage of the Pascual-Leone model is that in addition to accounting for cognitive development, it also focuses on the nature of the crucially important executive processes, postulating a wealth of theoretical concepts and processes. This is in contrast to the embarrassingly empty ellipse and the concept of the executive as a system for sharing resources between processing operations and temporary storage that has for too long characterized the Baddeley and Hitch model. However, while we initially chose to concentrate on the more tractable slave systems, work on the central executive has developed steadily, beginning with the proposal to adapt the Norman and Shallice (1986) supervisory attentional system (Baddeley, 1986), followed by further attempts to fractionate the executive. Again, we have attempted to keep

134 BADDELEY AND HITCH the model as simple as possible. We therefore proposed an executive that was purely an attentional control system, relying for storage on the slave systems and on LTM (Baddeley, 1996, 1998; Baddeley & Logie, 1999). The next stage was to attempt to fractionate the attentional control mechanism, postulating a hypothetical split into the capacity to focus attention, switch focus, and to divide attention across two concurrent tasks (Baddeley, 1996). Recent research has begun to provide evidence for such fractionation, having greatest success in the case of dual-task performance. Studies based on Alzheimer s disease suggest that this capacity is markedly impaired in contrast to normal elderly subjects, who retain the capacity to combine two tasks, provided the level of difficulty of the constituent tasks is matched across young and elderly subjects. Further investigation suggests that these results cannot be interpreted simply in terms of the level of overall difficulty of the various tasks, but seems to implicate a specific capacity for combining sources of information (Baddeley, Bressi, Della Sala, Logie, & Spinnler, 1991). A similar conclusion was reached by Bourke, Duncan, and Nimmo-Smith (1996) on the basis of studies using normal subjects. Finally, functional imaging studies are beginning to appear that seem to implicate the right frontal lobes in this function (D Esposito et al., 1995; Elliot, Dolan, & Frith, 2000). There are, however, other data that do not readily fit into the existing Baddeley and Hitch framework. For example, memory span for sentences tends to approximate 16 words, in contrast to a span of about 6 for unrelated words. Data from neuropsychological patients argue strongly against the assumption that this simply represents the additive combination of STM and LTM contributions, pointing to some form of symbiotic interaction (Baddeley, Vallar, & Wilson, 1987). A related problem is raised by a subset of amnesic patients who, despite grossly impaired LTM, can nevertheless perform well on the immediate recall of a prose paragraph or story that extends well beyond the capacity of the phonological loop. A similar capacity for excellent temporary retention of a matter of at least minutes is demonstrated by Clive Wearing, a densely amnesic patient with remarkably well preserved musical skills, who can obey the repeat signs while playing a harpsichord piece, moving on appropriately after the first repeat (Wilson, Baddeley, & Kapur, 1995). Another example is a densely amnesic patient described by E. Tulving (personal communication) who retains the capacity to play bridge, remembering the contract and even the cards that have been played earlier in the hand. All of these cases point to some form of storage that involves the temporary activation of long-term representations in order to create and maintain novel cognitive structures, referred to as long-term working memory by Ericsson and Kintsch (1995). Performance seems to depend on maintenance on the basis of ongoing attentional resources, but raises the question of what processes are involved in combining the activated representations into a coherent episode. In order to account for the foregoing observations and related phenomena,

REFLECTIONS ON KEMPS ET AL. 135 Baddeley (in press) has postulated a further component of working memory, namely the episodic buffer. This is assumed to act as a temporary storage system capable of holding information from the slave systems of working memory and LTM in some form of multimodal code. It is assumed to be controlled by the central executive, and to allow access through conscious awareness. As such, the mechanism is assumed to represent an approach to the binding problem, that is, the problem of how information from different sources is combined to create the perception of a single coherent episode. Postulating a new component after 25 years does not solve the deep and important problems underlying the issues tackled. It does, however, focus attention on the need for our working memory model to be able to account for the integration of information from multiple sources, and for the use of such integrated representations for building mental models and using them to plan for future action. In doing so, it might be seen as making our model somewhat closer to that of Pascual-Leone. For present purposes, however, we suggest that the two approaches should continue to develop alongside each other, in the hope that we will, in due course, reach a common solution, which, being based on different methods and sources of evidence, is likely to be all the more robust. REFERENCES Atkinson, R. C., & Shiffrin, R. (1968). Human memory: A proposed system and its control processes. In K. W. Spence (Ed.), The psychology of learning and motivation: Advances in research and theory (Vol. 2, pp. 89 105). New York: Academic Press. Baddeley, A. D. (1986). Working memory. Oxford: Clarendon. Baddeley, A. D. (1996). Exploring the central executive. Quarterly Journal of Experimental Psychology, 49, 5 28. Baddeley, A. D. (1998). Recent developments in working memory. Current Opinion in Neurobiology, 8, 234 238. Baddeley, A. D. (in press). The episodic buffer: A new component of working memory? Trends in Cognitive Sciences. Baddeley, A. D., Bressi, S., Della Sala, S., Logie, R. H., & Spinnler, H. (1991). The decline of working memory in Alzheimer s disease: A longitudinal study. Brain, 114, 2521 2542. Baddeley, A., Gathercole, S., & Papagno, C. (1998). The phonological loop as a language learning device. Psychological Review, 105, 158 173. Baddeley, A. D., & Hitch, G. J. (1974). Working memory. In G. H. Bower (Ed.), The psychology of learning and motivation Vol. 8 (pp. 47 89). New York: Academic Press. Baddeley, A. D., & Logie, R. H. (1999). Working memory: The multiple-component model. In A. Miyake & P. Shah (Eds.), Models of working memory: Mechanisms of active maintenance and executive control (pp. 28 61). Cambridge, UK: Cambridge Univ. Press. Baddeley, A. D., Papagno, C., & Vallar, G. (1988). When long-term learning depends on short-term storage. Journal of Memory and Language, 27, 586 596. Baddeley, A. D., Thomson, N., & Buchanan, M. (1975). Word length and the structure of short-term memory. Journal of Verbal Learning & Verbal Behavior, 14, 575 589. Baddeley, A. D., Vallar, G., & Wilson, B. (1987). Sentence comprehension and phonological memory: Some neuropsychological evidence. In M. Coltheart (Ed.), Attention & performance XII: The psychology of reading (pp. 509 529). London: Erlbaum. Bourke, P. A., Duncan, J., & Nimmo-Smith, I. (1996). A general factor involved in dual-task performance decrement. Quarterly Journal of Experimental Psychology, 49A, 525 545.

136 BADDELEY AND HITCH Brandimonte, M. A., Hitch, G. J., & Bishop, D. V. M. (1992). Verbal recoding of visual stimuli impairs mental image transformations. Memory and Cognition, 20, 449 455. Broadbent, D. E. (1958). Perception and communication. London: Pergamon. Brown, G. D. A., & Hulme, C. (1995). Modeling item length effects in memory span No rehearsal needed. Journal of Memory & Language, 34, 594 621. Burgess, N., & Hitch, G. J. (1992). Toward a network model of the articulatory loop. Journal of Memory and Language, 31, 429 460. Burgess, N., & Hitch, G. J. (1999). Memory for serial order: A network model of the phonological loop and its timing. Psychological Review, 106, 551 581. Case, R., Kurland, M., & Goldberg, J. (1982). Operational efficiency and the growth of short term memory span. Journal of Experimental Child Psychology, 33, 386 404. Chandler, M. J., & Chapman, M. (1991). Criteria for competence: Controversies in the conceptualization and assessment of children s abilities. Hillsdale, NJ: Erlbaum. Cowan, N., Day, L., Saults, J. S., Keller, T. A., Johnson, T., & Flores, L. (1992). The role of verbal output time in the effects of word length on immediate memory. Journal of Memory and Language, 31, 1 17. Daneman, M., & Carpenter, P. A. (1980). Individual differences in working memory and reading. Journal of Verbal Learning and Verbal Behavior, 19, 450 466. D Eposito, M., Detre, J. A., Alsop, D. C., Shin, R. K., Atlas, S., & Grossman, M. (1995). The neural basis of the central executive system of working memory. Nature, 378, 279 281. Elliot, R., Dolan, R. J., & Frith, C. D. (2000). Dissociable functions in the medial and lateral orbito-frontal cortex: Evidence from human neuro-imaging studies. Cerebral Cortex, 10, 308 317. Ericsson, K. A., & Kintsch, W. (1995). Long-term working memory. Psychological Review, 102, 211 245. Gathercole, S. E. (1995). Is nonword repetition a test of phonological memory or long-term knowledge It all depends on the nonwords. Memory and Cognition, 23, 83 94. Gathercole, S. E., Adams, A-M., & Hitch, G. J. (1994). Do young children rehearse? An individual differences analysis. Memory & Cognition, 22, 201 207. Gathercole, S. E., Frankish, C. R., Pickering, S. J., & Peaker, S. (1999). Phonotactic influences on short-term memory. Journal of Experimental Psychology: Learning, Memory & Cognition, 25, 84 95. Gathercole, S. E., Pickering, S. J., Hall, M., & Peaker, S. J. (2000). Dissociable lexical and phonological influences on serial recognition and serial recall. Quarterly Journal of Experimental Psychology, 53A, xxx xxx. Gathercole, S. E., Service, E., Hitch, G. J., Adams, A. M., & Martin, A. J. (1999). Phonological short-term memory and vocabulary development: Further evidence on the nature of the relationship. Applied Cognitive Psychology, 13, 65 77. Henry, L. A. (1991). The effects of word length and phonemic similarity in young children s short-term memory. Quarterly Journal of Experimental Psychology, 43A, 35 52. Henson, R. N. A. (1998). Short-term memory for serial order: The start-end model. Cognitive Psychology, 36, 73 137. Hitch, G. J. (1990). Developmental fractionation of working memory. In G. Vallar & T. Shallice (Eds.), Neuropsychological impairment of short-term memory (pp. 215 246). Cambridge: Cambridge Univ. Press. Hitch, G. J., Halliday, M. S., Dodd, A., & Littler, J. E. (1989). Development of rehearsal in short-term memory: Differences between pictorial and spoken stimuli. British Journal of Developmental Psychology, 7, 347 362. Hitch, G. J., & Towse, J. (1995). Working memory: What develops? In W. Schneider & F. E. Weinert (Eds.), Memory performance and competencies: Issues in growth and development (pp. 3 22). Northvale, NJ: Erlbaum. Hitch, G. J., Towse, J. N., & Hutton, U. (in press). What limits children s working memory span?

REFLECTIONS ON KEMPS ET AL. 137 Theoretical accounts and application to scholastic development. Journal of Experimental Psychology: General. Hitch, G. J., Woodin, M., & Baker, S. L. (1989). Visual and phonological components of working memory in children. Memory and Cognition, 17, 175 185. Hulme, C., Thomson, N., Muir, C., & Lawrence, A. (1984). Speech rate and the development of short-term memory span. Journal of Experimental Child Psychology, 38, 241 253. Jarrold, C., Baddeley, A. D., & Hewes, A. K. (1999). Genetically dissociated components of working memory: Evidence from Down s and Williams syndrome. Neuropsychologia, 37, 637 651. Kemps, E., De Rammelaere, S., & Desmet, T. (2000). The development of working memory: Exploring the complementarity of two models. Journal of Experimental Child Psychology, 77, 89 109. Macken, W. J., & Jones, D. M. (1995). Functional characteristics of the inner voice and the inner ear : Single or double agency? Journal of Experimental Psychology: Learning, Memory & Cognition, 21, 436 448. Morra, S. (2000). A new model of verbal short-term memory. Journal of Experimental Child Psychology, 75, 191 227. Neath, I., & Nairne, J. S. (1995). Word length effects in immediate memory Overwriting trace decay theory. Psychonomic Bulletin & Review, 2, 429 441. Nicolson, R. (1981). The relation between memory span and processing speed. In J. P. Das & N. O Connor (Eds.), Intelligence and learning (pp. 179 183). New York: Plenum. Norman, D. A., & Shallice, T. (1986). Attention to action: Willed and automatic control of behavior. In R. J. Davidson, G. E. Schwartz, & D. E. Shapiro (Eds.), Consciousness and self-regulation: Advances in research and theory (Vol. 4, pp. 1 18). New York: Plenum. Page, M. P. A., & Norris, D. (1998). The primacy model: A new model of immediate serial recall. Psychological Review, 105, 761 781. Pascual-Leone, J. (1970). A mathematical model for the transition rule in Piaget s developmental stages. Acta Psychologica, 63, 301 345. Smith, E. E., & Jonides, J. (1998). Neuroimaging analyses of human working memory. Proceedings of the national Academy of Sciences of the United States of America, 95, 12061 12068. Smith, E. E., & Jonides, J. (1999). Neuroscience Storage and executive processes in the frontal lobes. Science, 283, 1657 1661. Snowling, M., Chiat, S., & Hulme, C. (1991). Words, nonwords, and phonological processes: Some comments on Gathercole, Willis, Emslie & Baddeley. Applied Psycholinguistics, 12, 369 373. Towse, J. N., & Hitch, G. J. (1995). Is there a relationship between task demand and storage space in tests of working memory capacity? Quarterly Journal of Experimental Psychology, 48A, 108 124. Towse, J. N., Hitch, G. J., & Hutton, U. (1998). A reevaluation of working memory capacity in children. Journal of Memory and Language, 39, 195 217. Wang, P. P., & Bellugi, U. (1994). Evidence from two genetic syndromes for a dissociation between verbal and visuo-spatial short-term memory. Journal of Clinical and Experimental Neuropsychology, 16, 317 322. Wilson, B. A., Baddeley, A. D., & Kapur, N. (1995). Dense amnesia in a professional musician following Herpes Simplex Virus Encephalitis. Journal of Clinical and Experimental Neuropsychology, 17, 668 681. Received April 30, 2000; revised August 9, 2000