You are on page 1of 33

Annu. Rev. Psychol. 2012.63:1-29. Downloaded from www.annualreviews.

by California State University - Channel Islands on 09/12/12. For personal use only.

Working Memory: Theories,
Models, and Controversies
Alan Baddeley
by California State University - Channel Islands on 09/12/12. For personal use only.

Department of Psychology, University of York, York YO10 5DD, United Kingdom;
Annu. Rev. Psychol. 2012.63:1-29. Downloaded from


Annu. Rev. Psychol. 2012. 63:1–29 Keywords
First published online as a Review in Advance on attention, central executive, phonological loop, episodic buffer,
September 27, 2011
visuo-spatial sketchpad, short-term memory
The Annual Review of Psychology is online at Abstract
This article’s doi: I present an account of the origins and development of the multicom-
ponent approach to working memory, making a distinction between
Copyright  c 2012 by Annual Reviews. the overall theoretical framework, which has remained relatively stable,
All rights reserved
and the attempts to build more specific models within this framework.
0066-4308/12/0110-0001$20.00 I follow this with a brief discussion of alternative models and their rela-
tionship to the framework. I conclude with speculations on further de-
velopments and a comment on the value of attempting to apply models
and theories beyond the laboratory studies on which they are typically


Contents THE EPISODIC BUFFER . . . . . . 15
WORKING MEMORY: WM and Binding . . . . . . . . . . . . . . 15
THEORIES, MODELS, AND Visual Binding and WM . . . . . . . 16
CONTROVERSIES . . . . . . . . . . 2 Binding in Verbal WM . . . . . . . . 17
Short-Term Memory . . . . . . . . . . 4 LINKING LONG-TERM AND
Evolution of a Multicomponent WORKING MEMORY . . . . . . . 18
Theory . . . . . . . . . . . . . . . . . . . . 5 Is WM Just Activated LTM? . . . 18
CHARACTERISTICS OF THE Long-Term WM . . . . . . . . . . . . . . 18
PHONOLOGICAL LOOP . . . 7 LTM and the Multicomponent
The Phonological Similarity Model . . . . . . . . . . . . . . . . . . . . . 18
Effect . . . . . . . . . . . . . . . . . . . . . . 8 NEUROBIOLOGICAL
The Word Length Effect . . . . . . 8 APPROACHES TO
by California State University - Channel Islands on 09/12/12. For personal use only.

Articulatory Suppression . . . . . . . 8 WORKING MEMORY . . . . . . . 18
Annu. Rev. Psychol. 2012.63:1-29. Downloaded from

Irrelevant Sound Effects . . . . . . . 8 SOME ALTERNATIVE
Retaining Serial Order . . . . . . . . . 9 APPROACHES . . . . . . . . . . . . . . . 19
The Phonological Loop Theories of STM. . . . . . . . . . . . . . 19
and LTM . . . . . . . . . . . . . . . . . . 10 Theories of WM . . . . . . . . . . . . . . 19
The Phonological Loop: Cowan’s Embedded Processes
Master or Slave?. . . . . . . . . . . . 11 Theory . . . . . . . . . . . . . . . . . . . . 20
The Phonological Loop: Individual Difference–Based
Critique . . . . . . . . . . . . . . . . . . . 11 Theories . . . . . . . . . . . . . . . . . . . 20
VISUO-SPATIAL Jonides and the Mind
SKETCHPAD . . . . . . . . . . . . . . . . 12 and Brain of STM . . . . . . . . . . 21
Visual STM . . . . . . . . . . . . . . . . . . . 12 Computational Models
Spatial STM . . . . . . . . . . . . . . . . . . 12 of WM . . . . . . . . . . . . . . . . . . . . 21
Visuo-Spatial WM . . . . . . . . . . . . 13 WHAT NEXT? . . . . . . . . . . . . . . . . . 22
THE CENTRAL A Speculative Model
EXECUTIVE . . . . . . . . . . . . . . . . 13 and Some Questions . . . . . . . . 22
The Executive as Integration . . . . . . . . . . . . . . . . . . . . 23
Homunculus . . . . . . . . . . . . . . . 13 In Praise of Negative Results . . . 24
Fractionating the Executive . . . . 14 Applications . . . . . . . . . . . . . . . . . . . 24
Interfacing with LTM . . . . . . . . . 14 CONCLUSION . . . . . . . . . . . . . . . . . 25

WORKING MEMORY: THEORIES, since the early years, not all of course related
MODELS, AND CONTROVERSIES to or supportive of my own work, but a recent
attempt to review it (Baddeley 2007) ended
I was honored, pleased, and challenged by
with more than 50 pages of references. What
the invitation to write this prefatory chapter,
follows is a partial, as opposed to impartial,
pleased because it offered the chance to take a
WM: working account of the origins of the concept of multi-
broad and somewhat autobiographical view of
memory component working memory (M-WM) and of
my principal area of interest, working memory
M-WM: my own views on its subsequent development.
(WM), but challenged by the potential magni-
multicomponent My first draft would have filled the chapter
working memory tude of the task. The topic of working memory
page allowance with references; I apologize to
has increased dramatically in citation counts

2 Baddeley

is it cumulative or are we doomed to keep totally wrong. the newly developed digital computers. its director. Within psychology. a Viennese-trained philosopher that ignited what subsequently became known who argued strongly that a valid theory should as the cognitive revolution. Rev. time for the field of psychology. questions included. between Gestalt psychology and neobehavior.63:1-29. but for fields ing as it does postulates. where theory was en. I suggest learning in the albino rat. but I fully accept the importance of world. the issue. is psychology a science?.Channel Islands on 09/12/12. a very exciting Hull than to Tolman. Newton and Darwin got it right.” who no doubt stood in his attempt to produce a general theory of on the shoulders of lesser mortals like ourselves. be the case in philosophy? What would a good But what is the answer to our original ques- psychological theory look like? tion. allowing the work on optimizing the design of postal codes. by the time it was ism began to be challenged by new data and published.annualreviews. and Controversies 3 . For personal use only. to my own satisfaction at least. should theorists be architects. gradually extending bridge philosopher Richard Braithwaite (1953). in the published his seminal book. the whole field of learning theory new • Working Memory: Theories.annualreviews. ways myself am very much at the explorer end of the of organizing our existing knowledge of the continuum. It is an approach I have followed on asking the same questions. Perception and Com- United Kingdom at least. equations. providing tools both for interacting with the skills of the architect if theory is to develop. Donald Broadbent. learning. 2012. Edward My research career really began with my ar- Tolman in Stanford had a view of learning in rival at the Medical Research Council Applied rats that fitted this model. which had My own first published study (Baddeley benefited greatly from developments during 1960) attempted just such a crucial experiment. testable predictions. involv. I resolved at that Annu. if which would survive even if the theory proved so. as in the case of Darwin? Clearly both which scientific theories should aspire. which provided one of the sparks Popper (1959). that any complete theory is likely to require An alternative model of theorizing came explorers in its initial stages and architects to from Oxford. claimed that his success resulted from “standing nian model was explicitly copied by Clark Hull on the shoulders of giants. rived. where Stephen Toulmin (1953) turn the broad concepts into detailed models. all of those whose work should have been cited rival theories to confront each other in the and is turn led to a renewed interest in the philosophy point that if I myself were to develop a theory. the Second World War. laws. the study seemed to have collapsed. Neither side was able of control systems. and at a different stage of development. principally based on the study of maze Darwin had few such giants available. and As a result. Newton predictions. This approach was closer in spirit to sity College London in 1953. Its role Hull’s neo-behaviorist approach. Alas. that this was the case. with others influenced by to deliver a knockout blow. as appeared to ever since. championed by Cam. predicting that rats would be smarter than they riched by the need to tackle practical problems. using it to challenge Psychology Unit (APU) in Cambridge. Typical it would be based very closely on the evidence. This in abandoned the research area. prewar issues such as the conflict demonstrating. The first. all-important “crucial experiment” that settles I entered psychology as a student at Univer. the theory on the basis of more and more regarded Newton’s Principia as the model to evidence. and the year I ar- between the two apparently opposing views. Models. Downloaded from www. Psychol. building As students we were offered two answers to elegant structures such as Newton did. some based on cybernetics. had just The dominant answer to that question. the world and for further exploration. was provided by Karl munication. should they be explorers. This raised was to form a bridge between psychological the- the crucial question as to how you might decide ory and practical problems. www. of science as applied to psychology. and people simply by California State University . I argued that theories were like maps. I was assigned to make clear. or this question. should be according to Hullian theory. the Newto.

late. I went on to demonstrate that this pat- view of theory is the one that I continue to take. tern reversed when long-term memory (LTM) was required by using ten-word lists and several Short-Term Memory learning trials. 10% for similar) and a small but significant but also generating fruitful questions that will effect for semantic similarity (71% versus increase our knowledge. which implies a combination of free to explore this line of basic research. storage and manipulation. The Post Office thanked additive. and allegedly Popper himself. based system. semantic similarity then proved The term “working memory” evolved from the critical (Baddeley 1966b)..g. This more map-like 65%). sup).. It did not. Rev. pen. I will use STM to refer to the sim. in con. Psychol. STM: short-term memory ing memorable postal codes for each town in the effects of noise and similarity were simply the United Kingdom. response interference theory of verbal learning but that is another story. a short-term and the two are still on occasion used inter. however. which comprised a durable LTM com- was studying memory for proposed telephone ponent together with a temporary the ones I found interesting. memory errors resembled acoustic mis-hearing errors (e. on to a newly arrived colleague and I was left trast to WM. day. there were two storage systems. Downloaded from www. I found falsification. and 1 I subsequently abandoned the term “acoustic similarity” be- that memory for similar sequences (b g t p cause it suggested an input modality-based system. cat). Similarity was a they adopted could. Annu.Channel Islands on 09/12/12. old. good ). big. 4 Baddeley . soft. It was not intended as a statement of the concluding that STM depends on an acoustic linguistic basis of the memory system. measure of telephone line quality. pit. I tested this. (see Osgood 1949). not the case.63:1-29. tall ) with had subsequently abandoned the reliance on five dissimilar (wet. the code nitude of the similarity effect. So. at the APU in Cambridge and was prompted Other evidence came from amnesic patients by an applied problem. 2012. map. five dissimilar words (e. can. arguing instead that the mark of a (Baddeley 1966a) a huge effect of phonological good theory is that it should be productive. which led me to combine the classic tradition of I decided to see if the acoustic similarity ef- nonsense syllable learning with new ideas from fect could be used to provide sensitive indirect information theory. ilar sequences (huge. not similarity1 (80% sequences correct for dissimi- only giving an account of existing knowledge. resulting in my generat. mat. lar. v for b).g. dialing codes when he noted that even with A typical example of this was the recency effect visual presentation. I mistakenly assumed that phonological was a more neutral term. but the type of similarity By this time my approach to theory was seemed not to be regarded as important. wide. question. cow. comparing recall of sequences by California State University . A third source of test. which remains an open code (Conrad & Hull 1964).annualreviews. sible in situations that were far narrower than with five phonologically similar words (man. covered that within the philosophy of science. I subsequently dis. large. which is c) was poorer than for dissimilar (k r l q y). He tasks. and five semantically sim- Lakatos (1976). I concluded that earlier concept of short-term memory (STM). as is indeed the case in some countries. For personal use only. My telephony project was passed ple temporary storage of information. evaluating the quality of telephone lines that while other patients showed the reverse pattern might be more effective than a simple listening (Shallice & Warrington 1970). have been much central variable within the dominant stimulus- worse. My PhD supervisor Conrad had recently evidence came from two-component memory discovered the acoustic similarity effect. largely on the grounds of similarity in STM? that clear predictions only appeared to be pos. that of finding a way of who had preserved STM and impaired LTM. phonological and a long-term semantically changeably. evolving away from Popper’s idea of the need would Conrad’s effect generalize to other types for crucial experiments. but I was intrigued by the sheer mag- LTM: long-term memory me and went on their way regardless. I saw my work as fitting into a pattern of My interest in STM began during my time evidence for separate STM and LTM stores.

its assumption that of unrelated words (Baddeley & Levy 1971). in 1972. and the fact that we another ran a shop and a family. to LTM. We did www. One patient.63:1-29. Psychol. Patients with a digit span and semantically (Baddeley & Ecob 1970) and of only two items and an absence of recency in that standard tasks such as immediate serial re. given that Atkinson and Shiffrin short. playing an important general role semantic coding. After a first degree by California State University . where. and very effective for storing se. and recency should be Theory acoustically based. Material tested after a brief delay At this point. attentionally un. investigate the link between STM and LTM. but our own work showed no evi. that we would try to manufacture our own age system. Rev. Models. whereas Craik & Lockhart (1972) rial order. and operate over brief delays. In the case of two-component tasks. Downloaded from attributed their previously observed deficit to in Sussex and a PhD with Broadbent at the additional executive problems. the latter may take longer intellectual deficits. quent years. free recall should. In case. information in STM would guarantee transfer demanding.annualreviews. We had proposed (perhaps unwisely) to Korsakoff ’s syndrome. with STM phonologically based and and some form of temporary storage. Amnesic patients should have seman. I was joined by et al. but it has often been neglected in subse- not after a brief filled delay. For personal use only. may be based on either phonological or memory. Second.. Evolution of a Multicomponent tic coding problems. 2012. more elaborate processing leading it is much harder to use for storing the order to better learning. he had done a psychology MSc Annu. I moved to tients suffering from Korsakoff ’s syndrome did Sussex into a new department of experimental suggest a semantic encoding deficit (Cermak psychology.e. This point was made by Waugh “patients” using student volunteers. to phonological learning also occurs. retention of material over a brief assumed their short-term store to be a working period. semantic memory and levels of processing. was an efficient secretary. They did not. have a defective short-term store that components. Phonological coding First. unlike earlier items. each of which may be influenced should lead to impaired LTM. it be. Studies based on amnesic pa. meaningful sequences such as sentences. After nine years at the APU. the short-term store was essential for access to We also showed that word sequences can si. an STM task) is likely to reflect both LTM stores. often found in APU. and later work (Cermak & Reale 1978) in physics.Channel Islands on 09/12/ • Working Memory: Theories. to set up but tends to be more durable. LTM semantically based. Semantic coding can be rapid for showed that the nature of processing is crucial. according to Atkinson and call can reflect both long-term and short-term Shiffrin. It is worth emphasizing the need to distin. such patients should have major but readily forgotten. The former is easy to set up in cognition. Both can for instance. Third. This was not the by either phonological or semantic factors. was strategy dependent. LTM proved to be inconsistent with neuropsy- multaneously be encoded both phonologically chological evidence. 1970). beginning our grant just when the previously came clear that recency did not depend on ver. 1974). STM. Graham Hitch and I did not have access guish between STM as a label for a paradigm in to these rare but theoretically important which small amounts of information are stored STM-deficit patients and instead decided over brief delays and STM as a theoretical stor. the last few words & Norman (1965) and by Atkinson & Shiffrin of a list are well recalled on immediate test but (1968). my simple assumption of two (i. but with deeper. Graham Hitch as a post-doctoral fellow on dence of such a deficit (Baddeley & Warrington my first research grant.annualreviews. popular field of STM was downsizing itself bal STM (Baddeley & Hitch 1977) and that following criticism of the dominant Atkinson the use of semantic or phonological coding & Shiffrin (1968) model for three reasons. led to some clear pre- dictions. and Controversies 5 . the model assumed that merely holding of verbal material is rapid. in free recall (Glanzer 1972). Interest in the can learn new words indicates that long-term field began to move from STM to LTM.

However. approaching completion when I realized that We aimed to keep our proposed system as I had said nothing about the CE. abandoned the assump. similar results in studies investigating both Over the next decade we continued to verbal LTM and language comprehension. should we perhaps wait? We went ahead whatever system underpins digit span. and later adopting the verbal recall of sequences of spoken digits. or both. component the “visuo-spatial sketchpad. the disruption was far from from Gordon Bower to contribute a chapter to catastrophic: around 50% for the heaviest load. participants performed We began by focusing on the phonological a visually presented grammatical reasoning loop on the grounds that it seemed the most by California State University . given the very Annu. Figure 1 Norman and Shallice proposed that action The original Baddeley & Hitch (1974) working memory model. all of which were limited in capacity. very much simple as possible. In one study. The one directly relevant article I could find (Norman & Shallice 1986) appeared as a chapter because of the difficulty of persuading a journal to accept it (Shallice 2010. the digits should age rather than rehearsal. but not anyhow (Baddeley & Hitch 1974).” after the short-term storage system to varying degrees. not by removing the relevant part of their systems. initially referring to the ver- CE: central executive that was likely to occupy the limited-capacity bal system as the “articulatory loop. brain. We termed the third occupy more and more of available capacity. plication beyond the cognitive laboratory. po. second. We decided to two sources: first. spatial. because of the crucial importance of age. but by functionally disabling it by We labeled the central controller as a “central requiring participants to do a concurrent task executive” (CE). subvocal rehearsal assumed to be necessary to The concurrent task we chose was serial maintain information. presenting a a crucial one. most were concerned with the role of attention in perception.63:1-29. This was shown in Figure 1 (Baddeley & Hitch 1974). task while hearing and attempting to recall tractable system to investigate. For personal use only. our model was far from com- therefore suggested a clear involvement of plete. Psychol.annualreviews. alas. this point I agreed to summarize our progress proposing instead the three-component system in a monograph (Baddeley 1986). an influential annual publication presenting re- and perhaps more strikingly. We found broadly and many publications later. 2012. We hesitated. is controlled in two rather separate ways. Our results ory. so. all too common with papers presenting new ideas. personal communication). Downloaded from www.” leav- with the result that performance on any ing open the issue of whether it was basically task relying on WM should be progressively visual. impaired. At time increased linearly with concurrent digit this point. but at the same time.Channel Islands on 09/12/12. and split attentional control from temporary stor. One is 6 Baddeley . I unexpectedly received an invitation load. My re- tentially capable of being applied across a wide luctance to tackle the executive stemmed from range of cognitive activities. a case of Hamlet without the prince. whereas the principal role of the CE was the attentional control of action. At tion that WM comprised a single unitary store. Rev. and explore the model and its potential for ap- on the basis of these. Response extensive earlier research on verbal digit sequences of varying length. As term “phonological loop” to emphasize stor- sequence length increased. Although there were a on separate verbal and visuo-spatial short-term number of highly developed and sophisticated theories of attention. Performance slows systematically model that is still not complete nearly 40 years but does not break down. the error rate cent advances in the area of learning and mem- remained constant at around 5%. its probable complexity. which earlier research suggested might rely its attentional capacity.

whereas the phonolog- quence of my rejection of Fodorian simplicity ical loop model can. the development of WM in children. What then are the essentials of the broad None of these studies proved to be conclusive. For personal use only. as suggested by Lakatos (1976). A second assumption is that this traditional approaches to interference theory system is not unitary but can be split into an ex. beginning with and then apply them to data from the literature the phonological loop. it PHONOLOGICAL LOOP might better be regarded as a simple theory. I regard the very rigid defini. leading in due forgetting in the STM system was based on course to more detailed modeling. We used existing results. These of interference. the supervisory attentional sys. and still WM that had occurred in the previous decade of WM are discussed in turn. for example. which responds to situations that theoretical framework rather than a precise are not capable of being handled by habit-based model that allows specific predictions.annualreviews. overwriting. but each tion and LTM. that attempted to pull together developments in be widely applied. in three areas: fluent reading. with somewhat more remote ing evidence from a range of different phenom- connections to other systems such as percep. based on well-learned habits or schemata. A conse. We opted to assume a process of trace capacity system that provides the temporary decay. at the time (see Baddeley 1976. This involves combin- across modules. its productivity in generating good. In the 1960s. in my own is useful to postulate a hypothetical limited. theory? The basis is the assumption that it a state of affairs that remains • Working Memory: Theories. and Controversies 7 . it was a by vocal or subvocal rehearsal. a map with many blank areas that I hoped would number of studies attempted to decide whether be filled by myself and others. to create a and storage systems that are tightly interlinked simple model that is based on the method of within the module and more loosely linked converging operations. either by displacement or by three components could be regarded as mod. each consistent with the model. together ules in the sense that they comprise processes with our own subsequent studies. If so. prised. individually explicable in other ways. ena. volved in the many controversies surrounding tive activities. Downloaded from www. The processes.annualreviews. then this provides valuable www. The proposed components Annu. is the assumption that each of these systems can manding little in the way of attentional con. not only on With some relief. be fractionated into subsystems and that these trol. one concerning speech and which in turn implies some unspecified form sound while the other is visuo-spatial. producing a book (Baddeley 1986) questions linked to empirical methods that can by California State University . using the criteria proposed earlier. storage systems. linking together existing knowledge and gether with a means of maintaining information encouraging further investigation. If none tion of modularity by Fodor (1983) as unhelpful of the competing interpretations are able to ex- and neuropsychologically implausible. 2012. opinion. and the effects of aging. We saw the phonological loop as a relatively in the sense of Toulmin’s idea of theories as modular system comprising a brief store to- maps. Although I tended to refer to our proposals as CHARACTERISTICS OF THE a model. a relatively loose tem (SAS). This source of control can be overridden by a My overall view of WM therefore com- second process. trace decay or interference (see Baddeley 1976). Psychol. al- ecutive component and at least two temporary though we did assume a limited-capacity store. de. chapter 5). Models. Rev.Channel Islands on 09/12/12. of a road on your normal route. partly on the basis of our results and storage and manipulation of information that is partly because it avoided the need to become in- necessary for performing a wide range of cogni. An example of this might be the activity will be linked to perceptual and LTM processes of driving a well-learned route to your office. in ways that require further investigation. plain the whole pattern. I incorporated the its capacity to explain existing data but also on Norman and Shallice model into my own con. coping with the closure success of such a framework should be based.63:1-29. tractable cept of a CE.

are more likely to reflect loss of item sion also removes the phonological similarity information and to show semantic and other effect for visually presented materials but not LTM-based effects. the disadvantage of being required to confront however.g. then preventing it should by California State University . which has to Cambridge. The simple way of expressing our Colle & Welsh (1976) required their partici- results was to note that people are able to re. 43–49). either a decay or interference interpretation of short-term forgetting. As expected. Pierre Salame. support.Channel Islands on 09/12/12. Its effect is principally on the storage of order information.g. suggesting that it is better regarded five syllables (e. For personal use only. an effect that was cay is also likely to continue during the slower independent of the loudness of the irrelevant spoken recall of longer words. are required to continuously utter a single ited set. item information may be Articulatory Suppression helped by similarity since it places constraints If the word length effect is dependent on on possible responses. when participants were re- tion curve (Baddeley & Larsen 2007).annualreviews. oppor- as a perceptual effect. resulting the spoken material disrupted performance on in more trace decay and poorer recall. Rev. university. 1975b). consonants. However. and I followed up and extended 8 Baddeley . pants to recall sequences of visually presented member as many words as they can articulate in digits presented either in silence or accompa- two seconds (Baddeley et al. Suppres- sequence. (Baddeley et al. (2006) have suggested that the effect is lim- of sequences of five words ranging in length ited to the recency component of immediate se- from one syllable (e. Indeed. with longer The claim that auditory presentation allows words taking longer and hence allowing more a phonological trace to be laid down despite time for trace decay. We studied the immediate recall et al. 2012. there was a close corre- spondence between word length and tend to minimize the need to retain item in. illustrating the value The Phonological Similarity Effect of combining a broad theoretical map while leaving more detailed modeling to be decided As described above. Studies using word such as “the. Jones performance. whereas written material We assumed that vocal or subvocal rehearsal needs to be subvocalized if it is to register. for instance. pen day hot cow tub) to rial recall.” performance drops and is open sets. This approach has the advantage of since faced challenge and counter-challenge potentially producing a robust model. for example. studies subvocalization. a French visitor evidence for time-based decay. Fortunately. when presentation is auditory (Baddeley et al. nied by white noise or by speech in an unfamil- We interpreted our data by assuming that iar language that they were told to ignore. different words for each equivalent for long and short words.63:1-29. Irrelevant Sound Effects tion time. logical loop will function equally well with nations for each individual phenomenon. Only longer words take longer to rehearse. thus leading to poorer suppression has recently been challenged. but it has (see Baddeley 2007. the general hypothesis of a phono- a range of different possible alternative expla. the visually presented digits. tuberculosis. that specifically attempt to investigate the loop eliminate the effect. We presented sound sources. Downloaded from www. We interpret this as suggesting that The Word Length Effect spoken material gains obligatory access to the phonological store. dication that phonological storage is involved. although this tunity. pp. refrigerator) and found that may be true for long lists. This is indeed the case Annu. hippopotamus. shorter lists show an performance declined systematically with word effect that operates throughout the serial posi- length.. this is regarded as an in- by further experimentation. quired to read out words of different lengths as rapidly as possible. 1975b). Psychol. was likely to occur in real time. For this reason. When participants formation by repeatedly using the same lim. Such de.. 1984).

ilar items bunched together. that of trying to reduce the neg- Wales were developing a very extended pro.g. which assumes that both digits and ir. which sistent with hypotheses that depended upon in turn acts as a further stimulus. and Controversies 9 . initial hypothesis came to be regarded as cen. 2000). fluctu..Channel Islands on 09/12/12. notably includ- vant digits had no more effect on digit recall ing language. indeed. This again is an effect that a salutary lesson in premature theorizing. the rassment when it was clearly demonstrated that latter part is reproduced correctly. Colle and Welsh. including. when it occurs. as Lashley (1951) points out. cial feature was that the irrelevant sound needed the similar items appeared to be just as liable to fluctuate. if one item by California State University . which then gives rise to errors on the www. is lost. and yet it is often the case that ing.. for example. in which case it ating tones ( Jones & Macken 1993). Psychol. similar items as when they were adjacent. irrele. however. The argument goes as follows: If one considers a sequence of six letters as a series of Retaining Serial Order pairs. (e. we put the experiment to one side. so pothesis. bisyllabic words than from monosyllabic words.g.g. This proved to be something of an embar. We did. then we know that the principal source of A typical memory span is around six or seven interference comes from similarity at the stim- digits. despite our subsequent withdrawal. then the chain is broken and subsequent Annu. 5 9 6). but ruptive than dissimilar items ( Jones & Macken this disruption.. Colle’s work. nature of the phonological loop effect. our from dramatic. Downloaded from www. whereby each item acts as (Baddeley 1968) and were in particular incon- a stimulus for the response that follows. demonstrating that visual STM forgotten. Similarly. more problematic. For personal use only. not because the digits themselves are ulus level. It was only relevant sounds are represented as potentially later. in which sequences of sounds than did nondigit words containing the same within words and words within sentences must phonemes (e. it less disruption of our monosyllabic digits from is far from easy to explain how this is achieved. be maintained. Like has some major potential problems. whereby the cru. one two replaced by tun woo). Retaining serial order is a crucial demand words and nonsense syllables. However. recall should fail. Models. dfvkpl ). we suggested an interpreta. but also by a range of ity would come from having two or more sim- other sounds. In order might prove possible to greatly minimize the ef- to account for their results they proposed the fect by alternating similar and dissimilar items “changing state” hypothesis. chaining. for a wide range of activities. The most obvious hypothesis is through the concluding rather too hastily that this suggested previously described mechanism of chaining that interference was dependent on phonolog. despite errors in the middle of a sequence.63:1-29. find slightly piano. that I The OOE-R hypothesis is not spelled out in de. However. this ical similarity (Salame & Baddeley 1986). 7 5 3 to the remembered sequence were no more dis. is typically far 1995. 2012. if irrelevant items that were phonemically similar an item is repeated within the chain (e. They recall of postal • Working Memory: Theories. Larsen et tion in terms of some form of mnemonic mask. It seemed plausible to me showed that STM was disrupted not only to assume that the principal effect of similar- by irrelevant speech. realized that our result had clear implications tail but would appear to assume that serial order for theories of serial order retrieval in general is based on chaining. but rather because their order is was disrupted to the same extent by irrelevant lost. when I was attempting to pin down the competing paths on a multidimensional surface. A third phenomenon appears to be even tral to WM. and skilled motor performance suggesting that interference was operating at a such as striking a ball with a bat or playing the prelexical level. through sequential associations.annualreviews. Jones (1993) coupled this with the to be forgotten when sandwiched between dis- object-orientated episodic record (OOE-R) hy. The results were disappointing.annualreviews. then the chain should be disrupted. Rev. was discovered when trying to solve a practi- Meanwhile Dylan Jones and colleagues in cal problem. Unfortunately. ative impact of phonological similarity on the gram of research on irrelevant sound.

Furthermore. with her capacity for learning to pair unrelated fering an interpretation of the irrelevant sound Italian words. that crops up in a range of different The past decade has seen considerable activ. do items. normal. If so. the effect that language impairment and in normal children. a store and a serial order link. At recall. highly shared among the sequence of items. they require vocabulary (e. or other forms of ordinal marking. comparing this two stages. icant role in the initial stages of vocabulary ac- Can models of serial order in verbal STM be quisition and is also linked to reading skills (see 10 Baddeley . found a function for the phonological loop. subsequent response (Osgood 1949). of much model the link between the phonological loop greater practical importance was my collabo- and long-term phonological learning (Burgess ration with Susan Gathercole. the similarity sandwich effect. Both of these approaches can handle sary for new long-term phonological learning. and so forth. When effect in terms of adding noise to this additional compared to a group of matched controls. Vallar. One class of (PL) serve.g.Channel Islands on 09/12/12. they reflect a single common system? I myself selves were the main source of error (Baddeley think it more likely that evolution has applied 1968). whereas in fact the similar items them. flower-svieti ). serial order. We then came up with the further repetition before going on to the next idea that her phonological loop might be neces- strongest. What function might the phonological loop approaches fall into two categories. a point Modeling serial order continues to be a at which all the normal participants had perfect very lively field with considerable interaction performance (Baddeley et al. Although the work with PV had a major tious. but her auditory digit span was only hypothesis of Page & Norris (1998). Downloaded from www. which are now starting to become more ambi. again not a great item is retrieved first and then inhibited to avoid evolutionary gain. 2006). other than making telephoning models assumes that items are associated with a easier (an unlikely target for Mother Nature)? by California State University . 2012. two items. both in children with specific interpretation of chunking. deficit in phonological STM. We had between proponents of the different models. PV. for example (castle-table). She had fluent language production sumes a limited capacity of excitation that is and comprehension.63:1-29. 1988). an explanation as to her capacity to learn native language pairs was why similarity between irrelevant and remem. Very briefly. Giuseppe hypothesis. 2009). as they do not de. Psychol. stage (Page & Norris 2003). which may be tempo. Hitch et al. This result has continued to present a the same solution to a problem.. A makes sentences so much more readily recalled series of studies showed that WM plays a signif- than scrambled words (Baddeley et al. We tested this by requiring her to learn Russian pend upon chaining. For personal use only. Her intellect was A second approach is typified by the primacy preserved. of.’s (2000) OSCAR cropped up when an Italian colleague.annualreviews. ity in the attempt to produce clearly specified computational or mathematical models of serial order retention. The first artificial sentences in which ambiguity could item is the most strongly activated. series of internal markers. whereas she failed to learn a single bered items is not important. the second only be resolved by retaining the initial part of slightly less. Rev. Burgess and Hitch are now attempting to influence on my theoretical views. maintaining challenge to models of serial order. invited me to help him to investigate as in the case of the model and its subsequent a patient. 2009). which as. with a number located within The Phonological Loop and LTM the phonological loop tradition. while a plored the role of the phonological loop in vo- further challenge being addressed lies in the cabulary learning. except for long. The opportunity to investigate this question Annu. with a very pure and specific refinement by Burgess & Hitch (1999. in which we ex- & Hitch ral oscillators as in Brown et al. Russian word after ten successive trials. domains. the strongest a long sentence until the end. We would generalized to visual STM? The answer seems therefore expect errors to follow the similar to be that they can (Hurlstone 2010).

dis- tion flows from LTM to the loop. work with PV led to a loop can also provide a means of action con- major development. Switching familiar sounding nonwords such as loddenapish leads to a substantial slowing. Annu. A modification of the original model to take account possibly because of the availability of a few of the evidence of links between working memory simple tools such as the phonological similar- and long-term memory (LTM). and LTM. et al.annualreviews. and we wanted (Gathercole 1995). Downloaded from www. Similar results have been obtained and further developed by Emerson & Miyake (2003). We used dual task methods. switching was cur for the visuo-spatial sketchpad. Baddeley et al. or in the switching condition.b). www. We used a very simple task in which par- direct link from the loop to LTM. to cue their responses. representing more relevant plus and minus signs were provided on permanent crystallized skills and knowledge. however. this first became obvious treat WM and LTM as separate though interre- during a series of studies of the CE. in this case lated systems. Rev. fluence immediate nonword recall. I had previously tended to trol. and in another condition. becoming increasingly clear that the At a theoretical level.annualreviews. and suppression effects. as well as the rupting the CE with an attentionally demand- reverse. 2012. a process that later becomes sub- vocal.” etc. easier than less alternate addition and subtraction. In my own case. leading to disrupted almost as much by articulatory sup- a revision of the original model along the lines pression as by the much more demanding ex- indicated in Figure 2. to English. borrow- dren (Gathercole & Alloway 2008. The importance of self-instruction had of course already been beautifully demonstrated by the great Russian psychologist Alexander Luria. such as contramponist. we referred to the treatment of WM deficits in school-age chil- loop and sketchpad as slave systems. who showed that children gradually learn to control their actions using overt self- instruction. The Phonological Loop: Critique The loop is probably the best-developed and Figure 2 most widely investigated component of WM. the response sheet. Models. making the write down a total. a crucial distinc- ecutive task. ity. and Controversies 11 . word length. Psychol. For personal use only. the suppression effect dis- appeared (Baddeley et nonwords that have a similar letter structure to to subtract 1. It formed the The Phonological Loop: basis of an extensive and successful application Master or Slave? of the M-WM theory to the identification and In formulating our model. 2004a. 1998 for a review). Furthermore. It is. Gathercole ticipants were given a column of single digits (1995) showed that existing language habits in- and required in one condition to add 1 and by California State University . Here. He went on to demonstrate the value of self-instructions in neuropsychological rehabil- itation (Luria 1959). When the rary activation. it seemed reasonable to ing verbal task and a task involving simple ver- assume that a similar state of affairs would oc- bal repetition. This suggests that informa- to know why.. It is. 2001). represented by a using a simple subvocal code of “plus-minus- series of fluid systems that require only tempo- • Working Memory: Theories.Channel Islands on 09/12/12. To our surprise.63:1-29. It became clear that people were tion is made between WM. The fact that the loop specifically concentrating on its capacity for task switch- facilitates new phonological learning implies a ing. Gathercole ing the term from control engineering.

ipants making a same/different judgment. Meanwhile. tained a similar result for remembering a point tory suppression. Posner & Keele (1967) pro- Its strength is that it can provide temporary se. Bill Phillips. It is a system was based on speed of processing letters. All result was based on switching from a visual to of these are language related. Downloaded from www. how. The analogy that comes to mind is that of the Although this could reflect the duration of the role of the thumb in our motor behavior: small. how. This hypoth. however. appears to under. of a name code in letter judgments. seded by a phonological code after two seconds. In contrast to these spa- region of two digits (Larsen & Baddeley 2003).Channel Islands on 09/12/12. presenting either an identical stimulus or one scribes as the standard hypothesis is far from in which a single cell was changed. It is important to note. that the same system. It was et al. Annu. slowly developing phonological code that then by California State University . pp. We tested retention 38). found a steady decline over time. 2010). There is. by more generally. 1971). Psychol. nameable. and one which a visual letter code appeared to be super- that. for only two seconds. VISUO-SPATIAL SKETCHPAD Interest in visuo-spatial memory developed Spatial STM during the 1960s. regardless of I have discussed the phonological loop thus whether we measured performance in terms of far as if it were limited to the storage of heard accuracy or reaction time (Phillips & Baddeley and spoken speech. with partic- adequate as a theory of WM or even as a gen. as experimenters. visual trace. nonlinguistic auditory information such as en. suggesting WM. 12 Baddeley . we ignore at our peril. perhaps because of easier question of whether the same system is used for maintenance by subvocal rehearsal. Neither of these ory store using matrix stimuli. but it was disrupted by an interpo. in that is extremely useful. similar conclusion. that accuracy declines systematically with num- ing interest in comparing language and mu. although there is grow. Phillips went on to investigate the visual mem- vironmental sounds and music. ber of cells to be remembered (Phillips 1974).3 to 9 seconds. ever. I agree that what Nairne de. A range of studies by Kroll et al. overrides the visual. 2012. the cost is something in the located in an open field. duced evidence suggesting a visual store lasting quential storage. using a process that is rapid.63:1-29. although approximately half the cells would be filled at as discussed elsewhere (Baddeley 2007. involves sequential presentation and recall. it could equally well reflect a more not essential. came to a pin memory for both lip-read and signed mate. by which he appears to refer to attempts A colleague. 35– random on any given trial. their method and requires minimal attention. only one very limited component of lated information-processing task. demonstrating topics is well explored.annualreviews. operating under using articulatory suppression to disrupt the use broadly similar constraints. widespread. a danger of exaggerating its importance. tial memory tasks. but very useful. We eral account of STM. that the Posner and Keele rial (see Rönnberg et al. We chose 5 × 5 matrices in which esis seemed to be attributed to myself. Dale (1973) ob- by combining visual presentation with articula. ever. sic and some indication of overlap (Williamson suggesting limited visual STM capacity. Rev. this work that influenced our initial concept of the visuo-spatial sketchpad. which is spatially based and 30 seconds. which raises the a phonological code. When its use in digit span is prevented some form of active rehearsal. (1970).org It appears to be this that Nairne criticized un- der the label “the standard hypothesis” (Nairne Visual STM 2002). when Posner & Konick The most frequently used clinical test of visuo- (1966) showed that memory for a point on a line spatial memory is the Corsi block-tapping test was well retained over a period ranging up to (Milner 1971). For personal use only. Nairne’s criticisms do not apply to WM over intervals ranging from 0. 2004 for a review). and I decided to test to account for a range of time-specified STM this using material that would not be readily effects purely in terms of the loop. However.

and pain. demonstrating that some im- a sequence of blocks. We number of different receptor cells capable of de- found that recall of the visuo-spatially codable tecting pressure.” although the Brooks task. mation from all of these sources converges on leading to the question of whether the under- the sketchpad is far from clearly established. the “inner scribe. Research by Smyth and colleagues STM to its role in visual imagery. Logie (1995. In half of the cases the sentences can haptic coding as used in grasping and hold- be encoded as a path through a visually pre- ing objects. to be capable of attentional focus. Downloaded from www. Furthermore. Touch itself depends on a tions were not readily encodable spatially. about two less than digit span. We found that the tracking still the precise nature of visuo-spatial rehearsal re- disrupted the spatial but did not interfere with mains unclear. This we tested The nature of rehearsal in the sketchpad is using a task in which blindfolded participants also uncertain. and my assumption that infor- terpreted this result in terms of the sketchpad. 2012.Channel Islands on 09/12/12. pattern span can be disrupted by Klauer & Zhao (2004) critically review this by California State University . the verbal task. series of investigations controlling for potential tion (Della Sala et al. There are now multiple demonstrations while the other is preserved. whereas the brightness judg- ment showed a slight tendency in the opposite direction. The number of blocks used a visual imagery mnemonic whereby two tapped is increased until performance breaks unrelated items are associated by forming an down. 1999). sentences was differentially disrupted by pur- We currently know very little about these as- suit tracking (Baddeley et al. The other half of the instruc- a tactile component. The participant views an array of nine blocks disagreed and set out to show that I was wrong. 2011) suggests a tracked a sound source (spatial but not visual) distinction between a “visual cache. again while performing rehearsal system. Another possible channel of member and repeat back a sequence of spoken information into the sketchpad comes from sentences. concurrent visual processing. a distinc- tion that is also supported by neuroimaging Visuo-Spatial WM evidence (Smith & Jonides 1997). between visual and spatial STM.annualreviews. of the dissociation of visual and spatial WM. Robert WM. Della Sala chair could be remembered as a cow sitting on a et al. using a modified version of the chair. cold. Our own has suggested a kinesthetic or movement-based studies used a technique developed by Brooks system used in gesture and dance (Smyth & (1968). and www. He attempts to imitate this. and vice versa. research moved from visual seems likely. Models. storage. with Corsi span typically being around image of them interacting. in which participants are required to re- Pendleton 1990). scattered across a test board. 1975a). cow and five. for example. The tester taps He succeeded. whereas Corsi literature before performing a very thorough Annu.” a tempo- or detected the brightening of their visual field rary visual store. and Controversies 13 .org • Working Memory: Theories. Although these results convinced me that The CE is the most complex component of the system was essentially spatial. For personal use only. who was working with me at the time. Within the original model it was assumed Logie. We in- pects of STM.annualreviews. which in turn is likely to involve sented matrix. and a spatial manipulation and (visual but not spatial). I return to the artifacts.63:1-29. THE CENTRAL EXECUTIVE tem was spatial rather than visual (Baddeley & The Executive as Homunculus Lieberman 1980). (1999). Rev. and the participant agery tasks were visual rather than spatial. heat. leading us to conclude that the sys. line drawings or indeed by simple patches of with some patients being impaired on one color. Yet further fractionation of the sketchpad During the 1970s. showed that visual pattern be disrupted by visual stimuli such as irrelevant span is dissociable from spatial Corsi span is more susceptible to spatial disrup. their results support the distinction visual-spatial distinction at a later point. Logie (1986) showed that this process can Phillips matrix task. vibration. lying store was visual or spatial. Psychol.

in contrast to tasks that require sim. quiring explanation. we chose to study a task ally demanding tasks will place specific demands involving alternating between simple addition on the CE. The third executive capacity we investigated Annu. this created a number of rate modalities: one verbal. Downloaded from www. sence of titration of level of difficulty. A sec. with the deficit they can be tackled one at a time. This and related tasks coupled with a rather small additional effect have proved to be a successful strategy for sep. how subsystems using different codes could spatial tracking. decision making. but rather as a marker of issues re. with felt there might be a specific control system. suggestions as to the basic set of executive what would any adequate executive need to be capacities that are too numerous to discuss in able to do? I came up with four suggestions this context (see. I speculated as to what these might be. at the same time (Riby et al. however. We assumed to load the executive. when an executive load accompanied suppres- arating out contributions of the three initially sion. 14 Baddeley . tasks when performed singly. 2000. 2004). First it would need to be able 2004. Rev. One concerned the question of digit sequences. Psychol. by California State University .annualreviews. Participants do not simply use either tients were performing at the same level as both one code or another. We selected two tasks involving Much of our work has used concurrent tasks involved switching between tasks. Attneave (1960) I regard homunculi as poten. The study of task switching has expanded proposed WM subcomponents (e. arguing against a Fractionating the Executive unitary executive capacity for task switching.Channel Islands on 09/12/12. Although our model tended mance of the patients when compared to either to be criticized for taking this approach. ity. Miyake et al. storage. involving recall of problems. becoming theoretically rather complex. In an attempt to specify the functions of the I should point out that there are many other CE. sioned off. the assumption typically being that attention. be integrated without some form of common culty for each of these to a point at which our pa. It is impor. virtually a homunculus. to focus attention. However. a little young and elderly controls.63:1-29. 1999). and subtraction. focusing on Alzheimer’s with no storage capacity (Baddeley & Logie disease. As mentioned earlier. 1996). using a demanding concurrent ple maintenance. whereas simply found a large effect of articulatory suppression repeating 271 would not. 2004). capable of doing all the clever tracking and digit recall to operate simultane- things that were outside the competence of the ously.. We then required man in the head. In an attempt to constrain our targets or stimulus streams. The fourth executive task that I assigned to ond desirable characteristic would be the capac.g. like of the two control groups. Perhaps surprisingly. Baddeley very substantially in recent years (Monsell et al. hopefully in even greater when the two tasks are performed due course allowing the homunculus to be pen. the first place (Logie et al. Shallice 2002). For personal use only. but rather combine them. 2012. We titrated the level of diffi. For example. WM model. Provided the various jobs performance tends to decline with age on the performed by the homunculus are identified. Engle & Kane (Baddeley 1996). we had made the assumption ing this in collaboration with Italian colleagues that the CE was a purely attentional system for a number of years. In the ab- planation. 2005). verbal executive task and articulatory sup- ward in threes from a number such as 271 is pression as its nondemanding equivalent. counting back. evidence of this came from the impact of reducing attention on complex Interfacing with LTM tasks such as chess (Robbins et al. age did not disrupt this specific executive capac- tially useful if used appropriately. There was a marked deficit in the perfor- two subsystems. our homunculus was the capacity to interface ity to divide attention between two important with LTM. provided the level of difficulty is equated in tant that they are not seen as providing an ex. and in my view at least. for which we to disrupt the various components of WM. I had been study. for example. 2011). and the other requiring visuo.

we assume that a central role indicated by its name. years. 2000).. but also linking phrase can show the influence of phonological WM to perception and LTM. and Controversies 15 . Psychol. it is episodic in that of the buffer is to provide a multidimensional www. However. THE EPISODIC BUFFER WM and Binding The characteristics of the new system are Like Baars (1988). if it is to be issues. of the interaction between LTM and WM.Channel Islands on 09/12/12.63:1-29. Engle WM. Although it to ask interesting and tractable questions. In doing This capacity is particularly marked in the so. I was reluctant to add further systems to the a challenge that has kept Graham Hitch. providing rather than a simple additive effect (Baddeley a link with our earlier research on the vividness by California State University . 1987). He uses the WM stemming from the initial demonstration metaphor of a stage on which the products by Daneman & Carpenter (1980) of a corre. For personal use only.annualreviews. Models. but embarrassing for a model that had et al. Cowan 2005. the no potential for storage other than the limited concept appears to have been welcomed and capacities of the visuo-spatial and phonological is frequently cited. increasing to 15 when the words make up with Cowan (2005) in assuming a capacity in a sentence. This and similar tests that tems. It is able to coding at short delays and semantic coding do this because it can hold multidimensional at longer intervals (Baddeley & Ecob 1970). become lation between a measure they termed “WM available to conscious awareness. the region of four chunks.g. I felt that one in Richard Allen. theoretically productive. which assumes that consciousness purely attentional executive came from the serves as a mechanism for binding stimulus very extensive work on individual differences in features into perceived objects. where a single the components of WM. Such results were gratifying in with analyzing the executive and attentional demonstrating the practical significance of aspects of WM (e. a conclusion that is consistent with of visual and auditory imagery (Baddeley & Annu. the audience. with both visual and phonological codes being it is assumed to hold integrated episodes or usable simultaneously (Logie et al. it acts as a buffer • Working Memory: Theories. 2012. there is a need to use the episodic buffer (Baddeley 2000). I made the further based sequences seems to reflect an interaction assumption that retrieval from the buffer oc- between phonological and semantic systems curred through conscious awareness. but like most buffer stores it Memory span for unrelated words is around has a limited capacity. This enhanced span for sentence. On this point we agree 5. multicomponent theory. I decided to add a fourth component. and myself busy over recent 25 years was perhaps acceptable. 1999). representations. It had a number of advantages in addition require the combination of temporary storage to providing a possible answer to the question and processing have proved enormously suc. chunks in a multidimensional code. Andrade 2000). not only between case of language processing. et al. the actors. Our new component could be regarded as a Their measure required participants to read fractionation of our initial 1974 version of the out a sequence of sentences and then recall the CE into separate attentional and storage sys- final word of each. 1999 for in North America. Perhaps for this reason. 2009). which were more concerned reviews). At a cessful in predicting performance on cognitive theoretical level it formed a bridge between our tasks ranging from comprehension to complex own bottom-up approach based on attempting reasoning and from learning a programming to understand the peripheral systems first. Rev. Downloaded from later dual-task studies (Baddeley et al. of preconscious processes. In response to these and related suggests that people find it useful.annualreviews. and language to resisting distraction (see Daneman the more top-down approaches predominant & Merikle 1996 and Engle et al. span” and capacity for prose comprehension. This results in a theory of But how might this interaction occur? consciousness that resembles that proposed by A further challenge to the concept of a Baars (1988). although that subsystems.

Visual Binding and WM Our work on visual binding was strongly in- fluenced by some new developments that were beginning to extend the methods applied to the study of visual attention to the subsequent short-term storage of perceived items. A number of im- sources to be bound into chunks or episodes. for example. Vogel et al. so we speculated that the two had been bound together as a red that the buffer would depend heavily on the square (Vogel et al. If. but also attention-demanding process. component.63:1-29. Presentation of the stimulus array was accom- ing concurrent task should have a very substan. the CE controls access to and pothesis using our concurrent task procedure. whether our elephant would be or shape. observation that capacity was limited to about ing us to imagine something Figure 3 jects. In the initial (Baddeley 2000) model (see by Wheeler & Treisman (2002) obtained the Figure 3). We decided to examine bind.Channel Islands on 09/12/12. notably including the not only perceptually but also creatively. panied by a demanding task such as counting tial effect on the capacity to bind information. backward by threes. fol- lowed (after a brief delay) by a probe stimulus. We could regardless of whether participants were remem- then reflect on this new concept and decide. an ice-hockey-playing elephant. or were required to bind the two fea- better doing a mean defensive body check or tures and remember not only that a red stim- keeping goal. However. This all seemed likely to be an ulus had been presented. for example. As in the work of Phillips (1974). gesting that maintaining the binding of features using our well-tried concurrent task strategy to was attentionally demanding. I intentionally required all access same result when testing involved a single probe to go through the executive. as our initial hypoth. then an attentionally demand. and bound into perceived and remembered ob- Annu. from the buffer. on one hand. For personal use only. A central question of this approach concerned the fac- tors that determine the conditions under which features such as color and shape are integrated by California State University . should prove more detrimental to the bind- ing in two very different modalities. remembering a red square) binding of visual features into perceived objects than to either of the single-feature probe tasks 16 Baddeley . they found a binding impair- could then investigate empirically whether ment when the memory test required search- other links were needed. bering only a single feature. If the CE is heavily in- in contrast to minor effects from disrupting the volved in binding. Psychol.annualreviews.. then the concurrent task other subsystems. or a square. ing through an array of stimuli in order to find We studied the role in binding played by a target match. color for example. The basic experimental paradigm was de- The model following the introduction of a fourth veloped by Luck & Vogel (1997. with participants deciding whether or not the medium that allows features from different probe had been in the array. A subsequent study CE. disrupt each in turn. We ourselves tested the attentional hy- esis proposed. a system for 2001). namely the ing condition (e. the episodic buffer. it in- integrating information from a range of sources into a multidimensional code (Baddeley 2000). arguing that we item. volved presenting an array of visual stimuli. Downloaded from www. four objects and was approximately the same. and the binding of words into sen- tences on the other. a result they interpreted as sug- each of the three initial components of WM. portant results emerged. 2001). 2012.g. Rev. allow.

In conclusion.63:1-29. course. 2012. 2009). although binding is some- times discussed as if it were a unitary func- tion. It remains important in that it al- tween presentation and test. However. related word sequences. (e. sual binding. For- ticulatory suppression. tion is (see Baddeley et al.. results of which can be summarized quite simply for example. In one case. However. although concurrent tasks impair tial effect. Most importantly. study. which suggests that al. obtain a differen. Concurrent tasks involv- to which each shape should be bound were ing the visuo-spatial sketchpad had a small but presented in separate locations. Hence. none of of performance. and it may be temporary. of the advantage in recalling sentences over un- The final experiment of the Allen et al. We explored this disruption effect the episodic in time. complex scenes. while a third experiment presented ponent. terfere with the binding process itself.Channel Islands on 09/12/12. The evidence from both visual and verbal tion of bound stimuli. we assume operates rela- were as before: no additional binding deficit. and Controversies 17 . the results the case of sentences. be argued that perceptual binding is ward counting condition to one involving ar. It could. tively automatically in LTM. be displayed. As expected. colored shapes were overall performance. In this study. in no case did we obtain a these tasks differentially disrupted the binding differential disruption of binding (see Baddeley of words into chunks as reflected in magnitude et al. as required to per- cate that the simple binding of color and shape form WM • Working Memory: Theories. In another significant effect on recall that increased when by California State University . the active manipulation of information within traction did. atypical in not requiring central resources. as in the binding is not itself attention demanding. 2011 for an overview). finding. Rev.annualreviews. 2011 for a review). for maintaining visual bindings.annualreviews. parallel series of experiments investigating the this was just as great for the single features as role of executive processes in the binding of for the binding condition. just as with vi- (2006) paper did. Downloaded from www. for example.g. our data indi.g. again using simultaneous presentation. earlier items did show poorer reten. words into chunks during retention of spoken A series of further studies explored this sentences. however. the features to be bound were separated they also had a visually based executive com- Annu. followed by a probe. we suggest that it differs depending on Binding in Verbal WM the specific type of binding involved. they do not appear to in- presented sequentially. bindings involving. Binding was differ. maintaining bindings against distrac. as part of our converging op- an overall impairment in performance when erations approach to theory. ing an important but essentially passive struc- but this time inserting a single additional item ture on which bindings achieved elsewhere can that participants were instructed to ignore be. the demanding binding conditions. which we now regard as be- further. We interpreted this as binding is thus inconsistent with the original suggesting that binding did not demand extra proposal that the process of binding involves attention but that maintaining it against dis. of of new information to its context in LTM. For ex- Although it appears that attention may be useful ample. shapes and the color patches (Baddeley et al. a patch of red) and greater effect that was further amplified when the associated shape verbally. lows executive processes to carry out further entially impaired even though participants were manipulation. we had pursued a accompanied by backward counting. Simple articulatory suppression had a one feature visually (e. This may in turn lead to further told to ignore the suffix. For personal use only. We compared the back. Psychol. or durable. however. phrases into integrated sentences or objects into manding. however. Models. the binding of though visual binding per se is not attention de. a www. Although some both suppression and attentional load were re- of these activities led to a lower overall level quired. we found tunately.. binding may be perceptual or linguistic. using other concurrent tasks and more We carried out a series of experiments. which in When the final item was probed. red or square).

2003).. language habits for rehearsal. is likely to de. between cognition and action. is capable of handling information in a range of tions within LTM as well as procedurally based modalities and stages of processing. I see WM as a complex inter- I think. For personal use only. Psychol. I agree. It was only when I tried to represent my case of Cowan’s (2005) model. given careful and thorough investigation. NEUROBIOLOGICAL Long-Term WM APPROACHES TO Ericsson & Kintsch (1995) proposed this con. Ruchkin et memory involves the activation of many areas 3. He and I would.63:1-29. an interface that pend on phonological and lexical representa.Channel Islands on 09/12/12. All of these boost memory performance. Brain damage can be seen scientifi- cally as producing a series of unfortunate exper- iments of nature. It seems likely that some of the misunderstand- vated LTM (e. the active system that is able to provide an interface simplest component of WM. ings confronting M-WM stem from the rather by California State University . going on to extend it to who may nonetheless show normal binding the use of semantic and linguistic knowledge to in WM (Baddeley et al. it is a way of re. 2012. but I cannot see any advantage in treating this as a different kind LINKING LONG-TERM AND of WM rather than a particularly clear example WORKING MEMORY of the way in which WM and LTM interact. result in bound these and other situations utilize previously de- representations accessible through the episodic veloped structures in LTM as a means of boost- buffer. I assume that in the LTM. expert mnemonists.annualreviews. and particularly by patients with specific im- pairment in the absence of general cognitive deficits. views in the simple model shown in Figure 4 ferring to those aspects of WM that are not his that we found we agreed. This is also true tween myself and Robert Logie. for which activated LTM is not that all information entered the sketchpad via taken as an explanation.g. however. Downloaded from www. linking the performance of 18 Baddeley . ing WM performance. who insisted of language. agree that the phonological loop. This was also reflected in a disagreement be- of the brain that involve LTM. Rev. Is WM Just Activated LTM? LTM and the Multicomponent Model A number of approaches describe WM as acti. These can then be ex- tended to help diagnose and treat patients with Figure 4 related but more complex disabilities. My current view of the complex and multiple links between working memory There are two aspects of such research: (WM) and long-term memory (LTM). My view on this issue is that working limited links with LTM shown in Figures 2 and Annu. 2010). Cowan 2005. Nature is not usually a good experimenter: Patients typically have a range of different deficits. WORKING MEMORY cept in explaining the superior performance of The development of my own views on WM has been strongly influenced by the study of patients with neuropsychological deficits. fluenced by LTM. They argue that types of binding may. the behavioral. capacity that is disrupted in amnesic patients. allow clear theoretical conclusions to be drawn. Incoming information current principal concern and not a denial of a is processed by systems that themselves are in- need for further explanation. but just occasionally “pure” deficits occur that potentially.

There are some important areas. which such as trace decay. Rev. My own view was of the temporal structure of activation than is that it probably formed part of a complex cir. The second more ba- the frontal cortex was associated with success. typically available at present. and the neurobiological basis of WM. of awake monkeys performing a visual STM doubt. the patient to cognitive psychology. My own view is that this simply of short-term storage. my own expertise and current concern is which offers an essential and potentially pow- for the behavioral and the extent to which erful tool for understanding cognition and its neurobiological research has so far contributed neural basis. are concerned with the specifics of the model.63:1-29. see M-WM concerns those aspects associated with Henson 2001. APPROACHES A second source of apparent support for the Theories of STM concept of WM came from neuroimaging when the various subsystems of M-WM appeared to A good deal of the controversy surrounding be relatively closely localizable (for reviews. Single-unit recording of the brains of WM. sic concern is the validity of the assumption that ful recall. and Controversies 19 . However. cuit underpinning the visuo-spatial sketchpad. as well as the need to mize semantic and other longer-term factors. who comment on the inevitably include long-term factors such as se- “near-revolutionary changes in psychological mantic coding and will emphasize the similar- theories about STM. That means using tasks that mini- the complexity of WM. and in particular with to a large number of further neuroimaging in.Channel Islands on 09/12/12. however. In the case locations. The first concerns the lack of apparent Annu. This led STM rather than WM. anatomical localization will provide a firm the- ated with failure (Goldman-Rakic 1988).annualreviews. This oretical basis for a system as complex as WM led some to conclude that it reflected a spe. Smith & Jonides 1997).org task found that continued activation of cells in replicability in the field. ities between STM and LTM. (2008). own work has focused on analyzing one source chapter 7).org • Working Memory: Theories. linking it to its anatomical and vinced that it has made a major contribution neurophysiological basis. Their very ambitious in- terpretation of the “mind and brain of STM” Theories of WM is discussed below. 2012. ponent in testing and potentially validating dis- logical studies that appear to suggest that WM tinctions such as that between “remembering” may depend on one or more specific anatomical and “knowing” (Düzel et al. with similarly great ad. modify paradigms to fit the constraints of neu. for example. in the past decade. These will Jonides et al. not reflect a general rejection of neuroimaging.annualreviews. as in the case A different interpretation is offered by of Nairne’s (1990) feature model. although There are a number of ambitious models of I think it is very important to understand the WM that I regard as broadly consistent with www. producing a range of different results. Downloaded from www. other concerns when fed into a meta-analysis often failed to stem from a difference in purpose. There is no doubt that the popularity of the in which neuroimaging provides a crucial com- concept of WM owes a great deal to neurobio. the phonological loop model. to the cognitive understanding of M-WM. I am not yet con- neurobiological. I regard these vances in the neurosciences” that have occurred two approaches as complementary. and disrupted activation was associ. Other theorists focus on the whole range of roimaging. such as those investigating conscious awareness. codes that may be contributing. Subsequent discovery of similar cells elsewhere SOME ALTERNATIVE in the brain is consistent with this view. in the absence of a much better understanding cific frontal location of WM. For personal use only. 2001). This does and will ultimately be combined. that based on phonolog- reflects the unreliability of such results and ical coding. Models. avoiding overt speech. for example. Psychol. Whereas some vestigations from many different laboratories. More generally. Both are important to psychological theories of WM. Much of our show a consistent pattern (see Baddeley 2007. I have two major sources of by California State University .

Individual Difference–Based Theories I suspect that a theory of executive processing The demonstration by Daneman & Carpenter based entirely on inhibitory control may be a (1980) that WM-span measures can predict little narrow. They emphasize the importance of in- issues but differ in our terminology and areas hibitory processes. but comprehension has provided a major focus of I suspect people also differ in more positive as- research on WM over the past 30 years. there has been considerable inter. I see Cowan’s model as prin- to shielding the memory content from poten- cipally concerned. however. although each correlational approaches to this issue have has a different emphasis and terminology. the capacity is much Annu. Much of their work involves a link between the CE and the episodic buffer. Cowan ate items from a semantic category. research that interacts ference across tasks ranging from recall from with and is complementary to the phonological episodic LTM.. They have demon- of the processes operating beyond attentional strated that such groups differ not only in WM focus as reflected in his extensive and influen- performance but also in susceptibility to inter- tial work on verbal STM. Others specify the capacity of this attentional focus and also assume the need to resist time-based decay hence the capacity of WM. I regard our differences as princi- mance on an antisaccade task of eye movement pally ones of emphasis and terminology. Rev. those with very high and those accepting the need for a more detailed analysis with very low WM span. consistent with the broad M-WM framework est in identifying the feature of such com. the various tasks involved in span (Towse et 2000) or they emphasize the role of interference closer to four. each of which may contain However. control (for a review. two subgroups. unlike an earlier by California State University . through the capacity to gener- loop hypothesis of verbal STM (e. however.annualreviews. this is four rather than decay (Saito & Miyake 2004). Engle & Kane own. we agree on most 2004). which they argue are crucial of current focus. 62). Cowan’s theories pinning WM capacity is that proposed by Engle might seem to be totally different from my and colleagues (Engle et al. 1992). and in my view. a number of limitations. in my terminology. Some state” (Cowan 1999. the most promising work in this area comes Cowan’s Embedded Processes Theory from combining experimental and correla- tional methods to tackle the question of why Cowan defines WM as “cognitive processes some people are better able to sustain ma- that are maintained in an unusually accessible terial under these complex conditions. Importantly. typically initially testing a tem works as “activated LTM” but does not large group of participants and then selecting treat this as providing an adequate explanation. 2012. theoretical account of the mechanisms under- At a superficial level.63:1-29. He produces exten- but emphasize efficiency at switching between sive evidence to suggest that. the most extensively developed more than a single item (Cowan 2005). Downloaded from www. For personal use only. with the tial disruption. 2004). His theory involves explanations focus on the capacity to utilize a limited-capacity attentional focus that oper- gaps between the processing operations of ates across areas of activated LTM. 1999. Psychol. At a theoreti. combination of individual difference and exper- Cowan refers to the material on which his sys- imental approaches. of WM based on individual differences as cal level. see Engle & Kane 2004). In practice. in. I would see most of the models (Daneman & Merikle 1996). Control is clearly important. pects of attentional capacity. Although this is an impressive program of work. Purely and verbal STM components (see Alloway 20 Baddeley .g. suggestion of seven items. to perfor- et al. typically focusing on executive control but plex span measures that allows them to predict accepting the contribution of separate visual cognitive performance so effectively. chunks or episodes. volving multiple replications and extensions In general. p. the multicomponent framework.Channel Islands on 09/12/12. A central the span task in order to maintain a fading issue for Cowan over recent years has been to memory trace (Barrouillet et al.

org This approach ( Jonides et al. however. I remain somewhat patients have a general difficulty in binding fea. the conclusions are based on spatial binding. theorists who attempt a tient (Baddeley et al. 2008) is strongly tions about neurobiological processing levels. Potentially more powerful evidence tween primary and secondary memory. 2003). giving rise to a range Squire 2004). and LTM are all performed in the same over a two-second delay. using each of a range of modalities. 2012.g. but that does not mean and colleagues (Unsworth & Engle 2007) have that all these are necessary in order to repeat recently reverted to an earlier distinction be. the classic neuropsychological lit- the M-WM model (see Figure 2). developmental am. note WM. influenced by neuroimaging in assuming. For personal use only. for its phonological. whereas classic am. word is likely to activate regions responsible Once again. with my own preference cized on two grounds: first. However. skeptical as to how productive such a model will tures together (Hannula et al. lexical.Channel Islands on 09/12/12. p. where Barnard’s (1985) ambitious computation- STM tasks often activate areas of the brain ally based “interacting cognitive subsystems” that also are involved in LTM (e. much more detailed account of WM. of different models of WM. articulatory. This reflects a difference between 2006). 2010). and Controversies 21 . that the measures for the gradual development of detailed modes used comprise both long. 2011). performs well on a range of aspects of M-WM including the sketch pad complex WM tasks (Baddeley et al. suggesting that amnesic ating across 10 neural levels. 1037) and the loop The major source of evidence cited by (Anderson et al.63:1-29. et al. Nonspatial binding such as that of The WM theorists discussed so far have color to shape was not found to be impaired all taken a broad-based approach to theory. which can on nesic patient Jon. whereas ponents (Shrager et al. However. Olson et al. that Jonides et al. Presenting a 1987). 2008). but the interpre- between the fluid and crystallized systems that tation of this has been questioned (Baddeley reflect temporary structural representations in et al. The “mind and brain” model proposed by Jonides and the Mind Jonides et al. erature typically reports a dissociation between perceptual and memory deficits (Shallice 1988). that perception.and short-term com. typically nesic patients do not appear to show evidence accompanied by computer simulation. There is strong evidence that both spatial pro- cessing and episodic LTM depend on the hip. It was initially developed to account for in their discussion of the single-unit studies. Rev. 2006). for They go on to illustrate their model. There are. this evidence has been criti. Models. who has greatly reduced hip. occasion result in subcomponents resembling pocampal volume.. is somewhat complex. They also cite evidence that involves 13 psychological processes oper- from neuropsychology. Jonides et al. Engle semantic dimensions. within a broad theoretical framework. 2010). Ruchkin model can also be mapped directly onto M- et al. a very flexible approach. Computational Models of WM • Working Memory: Theories. as Jonides et al. overall similarities may be ob. us in theoretical style. are rather more ambitious. Furthermore. comes from neuroimaging. language processing but was subsequently used the fact that an area becomes active during a extensively by Barnard to analyze situations in- given task does not mean that it is essential volving human–computer interaction (Barnard for performance on that task. and scured by terminological differences. The model can simulate most aspects www. 2006 for an example of such a model). resulting in a figure anatomical locations. involving and Brain of STM by California State University . Downloaded from www. 1996). which exists based on lesions in neuropsychological I would interpret in terms of the distinction patients (Olson et al. the case of remembering three visual items STM. 2006. in a hippocampally compromised amnesic pa.annualreviews. prove to be. This is of a WM deficit (Baddeley & Warrington 1970. 2004. (Anderson et al.annualreviews. Psychol. that word. and second. five psychological assumptions and six assump- Annu.

I am a theo- due course also prove to be a problem for an retical mapmaker and temperamentally skepti- ambitious new model proposed by Oberauer cal of complex theoretical architectures. For personal use only. (a) maintaining structural representations by I have described my attempts to turn a broad dynamic bindings. which the bridge of a ship. theoretical framework into a more detailed (c) flexibly reconfiguring them.63:1-29. for the whole WM system. namely. for achieving each of these.” will become Annu.” He proposes and Some Questions six requirements for a WM system. worked with Barnard in a simulation. In discussing his attempt to of this aspect of procedural working memory. components in turn. He sees the main focus of WM as being “to serve as a blackboard for information processing on which we can WHAT NEXT? construct new representations with little A Speculative Model interference from old memories. apart from the episodic buffer. In each case this suggests questions procedural WM. (e) controlling empirical exploration. We would expect this simple instruction for others to use. This is an attentional sys- sume that an example would be the process con.” as in broad. Downloaded from www. the evidence at model using components that can be built in present is rather sparse. a computer scientist who had increasingly influential. whereas proce. and insightful model. and ( f ) encoding new struc. and neither sheer complexity of the model makes it difficult for 3. who attempts to provide a blueprint will tell. However. and the A crucial feature of Oberauer’s model is the attempt to provide considerably more specula- distinction he makes between declarative and tive detail. tem. ambitious. (b) manipulating them. and what I myself would call in Barnard’s hands has been applied with the central executive. now This is certainly a very ambitious program. dural WM is concerned with the nondeclarative processes that underpin such operations: I as. appropriate to end on my own current specula- tures into LTM. (d ) partially model by a process of speculation followed by decoupling these from LTM. But then. he also capacity component of Cowan’s (2005) model? 22 Baddeley .org Bowman (2011). I think the investigation subcomponents. Time (2010). plexity of the model may make it difficult to I suspect that undue complexity may in evaluate experimentally. press the green for number 2. produce a full simulation of the model. hence attempting As Figure 5 shows. and levels of awareness as part of a through what he refers to as the “bridge. Howard sometimes referred to as “task set. However. Central executive. It is therefore perhaps LTM retrieval. It portant the computational detail really is. comprising most tially offer a way forward. the sheer com- premature attempts to specify at a neural level. He postulates mechanisms tions and some of the many questions they raise. the pears. my current views are not to put flesh on the previously vague concept of dramatically different from our original specu- “activated LTM. to be followed throughout the experiment. how does it differ from the limited- trolling subvocal rehearsal. advocates a hierarchical decomposition of the and as Oberauer points out. avoiding unnecessary detail such as citing development. Rev. and is as if some mini-program is set up and then indeed whether it gives an adequate account of runs. considers a higher level of procedural control emotion. by California State University . However.annualreviews. Psychol.” lation. It is also unclear how im. I consider the various of the current work in the area.Channel Islands on 09/12/12. but we currently know very little about what is happening within the more peripheral how this is achieved. of WM while linking it to motor control. Consider the following: success to an impressive range of situations A participant in my experiment is instructed to from choreography to theories of depression press the red button when the number 1 ap- (Teasdale & Barnard 1993). but it could be an ex- isolation. 2012. Declarative WM is the aspect that will not be easily answered but that poten- of WM of which we are aware.

Visuo-spatial sketchpad. our attempt to link the loop with LTM through ing. where within the M-WM system is this located? I assume that the buffer provides access to Integration conscious awareness. Can we reach a conclu- trol in terms of a single factor such as that of in. Rev. only of their products when registered in the utive functions.Channel Islands on 09/12/12. and how are buffer? they organized and interrelated? Just how far can one take attempts to explain executive con. does this mean that we are Finally. tactile. and Controversies 23 . Phonological loop. Psychol. is this true of hap- can other modalities such as smell and taste be California State University . For personal use only. LTM prove equally useful? nic detector system. Logie (1995) suggests. or is it more like atten- rent speculation is that it operates according tionally based refreshing? Finally. visuo-spatial • Working Memory: Theories. (2002). tial aspects appear to be clearly separable but ity between chunks? If similarity is important. how do these increasingly detailed not directly aware of the other subsystems but accounts fit together to provide an interactive www. linked within the sketchpad. given that to the principle of attentionally based Figure 5 A speculative view of the flow of information from perception to working memory. and kinesthetic memory? What is added without impacting visual or verbal capac. the mechanism of visuo-spatial rehearsal? Is it ity? Are there separate subsystems for smell and a spatial analogue to the phonological loop.63:1-29. as I suspect? To what between individuals? Do we need a concept of extent is the loop used for remembering non- cognitive energy? verbal material such as music or environmental sounds? Episodic buffer. How should we measure its capacity? To what extent is this limited by num.annualreviews. Downloaded from www. What language acquisition proved very fruitful. 2012. would about emotion? Elsewhere (Baddeley 2007) I pursuing the link between the sketchpad and have suggested that it impacts WM via a hedo. sion on the ancient trace decay/interference hibition? Do we need to worry about precisely controversy? Is subvocal rehearsal atypical of what is being inhibited and whether this differs other types of rehearsal. as discussed by Johnson et al. I assume that it comprises a number of exec. as taste? How is rehearsal maintained? My cur. VSSP. The visual and spa- ber of chunks and to what extent by similar. but how many. Models. Annu.

Rev. far too im. 2011). to understand WM. as an explanation of the differences between ity is addressed through inclusion of other con. M-WM framework to specific practical prob- ready know. ical framework together with the availability ploration of the episodic buffer (Baddeley et al. however. though limited. Another instance Annu. For personal use only. 2004). through direct application of the confirmation of what we are pretty sure we al. WM measure applied to school-aged children sults. be a good justification for the question. it is important to A second aspect of theoretical application know what it does not do. dimension and of dual-task performance as a sion after surveying the experimental literature way of controlling of whether the experiment has sufficient power is the development and validation of a dual-task and indeed whether the question is worth ask. unitary system that mediates between percep. Nega. the boundaries of the laboratory-based effects ever. 2009. First. provided the problem of sensitiv. Gathercole’s extensive development of a an avoidance of too much risk of negative re. they are hard to interpret. with our original 1974 studies. If we are et al. be very important and development of working memory is proposed publishable. and this is likely is the use of the M-WM theory as a tool for to involve negative results. This can to have had success in achieving this. it does appear much more difficult to answer them. Such techniques and its implications for clinical and social psy. control. areas. as any diver or singer will know. This was the case Coolidge 2010). where the tive results can. They also raise the question (Gathercole & Alloway 2008). First of all. 2012. but cussed this aspect of M-WM. suggested ditions showing positive effects together with by a study of surviving artifacts (Wynn & clear evidence of replication. wide range of problems. however. points. Although I have not dis- It is of course very easy to raise questions.annualreviews. Neanderthal man and homo sapiens. and there needs to neuropharmacology to language therapy. experimentation. How. extremely valuable both in helping explore portant to be left to working memory.63:1-29. Psychol. and lems. and even to paleoanthropology. My own view is that this breadth of applica- fect of concurrent tasks was much less than tion has reflected the simplicity of the theoret- anticipated. Downloaded from www. Applications tion. and even more so in our recent ex. Negative results are a pain for a number of has been successful in identifying children at reasons. Publication ranging from human factors to psychiatry. with all these factors mak. do at least provide concep- an overall cognitive system that attempts to tual and practical tools for investigating a minimize the demands made on WM while al. LTM. of similarity effects as an indication of coding So what does WM not do? My own conclu. 24 Baddeley . resulting in confirmation bias. Here the applications are very extensive. performance measure for the early detection of ing in the first place. of a few basic methodologies. have the potential to become future growing Suicide by breath holding is not an option. From a theoretical lowing it to intervene where necessary. at two lead to a program that cautiously seeks easy levels. risk and providing methods of helping teachers They could result from a poor design or sloppy identify and help children with WM problems by California State University . A very viewpoint. as has often proved investigating and understanding other research to be the case in the various stages of devel. Alzheimer’s disease (Kaschel et al. where the ef.Channel Islands on 09/12/12. such practical applications can be basic example is that of breathing. and while not guaranteeing chology (Baddeley 2007) is that we have evolved fruitful answers. oping the current M-WM model. is justifiably more difficult. such as the use 2009. we clearly and in highlighting theoretical anomalies that do have considerable. Logie ing negative results harder to publish. are easily learned. and action? A central requirement of our original frame- work was that it should be applicable out- In Praise of Negative Results side the laboratory.

63:1-29. J. Allen RJ. Andrade J. Rev. Allen RJ. ed. UK: Cambridge Univ. Q. Is the hippocampus necessary for visual and verbal binding in working memory? Neuropsychologia 48:1089–95 Baddeley AD. Psychol.Channel Islands on 09/12/12. 2004. Working Memory. 129:126– 45 Baddeley AD. The phonological loop as a language learning device. 18:302–9 Baddeley AD. 1966a. 2000. 1998. Vargha-Khadem F. 1970. Annu. and Christopher Jarrold for many stimulating discussions and for commenting on an earlier draft. Psychol. 1996. coupled with the Lakatos crite. J. 2012. 1966b. UK: Oxford Univ. Simultaneous acoustic and semantic coding in short-term memory. Thought and Action. pp. 30:221–56 Atkinson RC. 2000. 2010.: Gen. An integrated theory of the mind. Q. 77:1698–716 Anderson JR.: Gen. Lebiere C. J. J. 1986. els of the components of WM and their mode rion of judging success by productiveness rather of interaction. Nature 277:288–89 Baddeley AD. JT Spence. 111:1036–60 Anderson JR. In defense of homunculi. Human memory: a proposed system and its control processes. Shiffrin RM. to develop interlinked and more detailed mod- ories as maps. 1968. However. A 49:5–28 Baddeley LITERATURE CITED Allen R. 18:362–65 Baddeley AD. 1976. Psychol. Rev. Cogn. 2011. Psychol. Rev. Psychol. 89– 195. New York: Basic Books Baddeley AD. Exp. 73:454–57 Baddeley AD. 1996. CONCLUSION than predictive accuracy. Baddeley AD. Is the binding of visual features in working memory resource- demanding? J. 105:158–73 www. Exp. In Sensory Communication. ACKNOWLEDGMENTS I am grateful to Graham Hitch. Cambridge. 2007. Short-term memory for word sequences as a function of acoustic. Working Memory. it hopefully will be increasingly possible preference has been for Toulmin’s view of the. Enhanced learning of a position habit with secondary reinforcement for the wrong response. Models. 1968. Susan Gathercole. New York: Academic Attneave F. Press Baddeley AD. Psychol. 777–82. W Rosenblith. Gathercole SE. A Cognitive Theory of Consciousness. 20:249–64 Baddeley AD. Sci. UK: Oxford Univ. 1960. KW Spence.annualreviews. Verbal and visuospatial short-term and working memory in children: Are they separable? Child Dev. For personal use only. Robert Logie. Exploring the central executive. 4:417–23 Baddeley AD. 135:298–313 Alloway TP. Neuropsychologia 49:1393–400 Baddeley AD. Lebiere C. my own map. Reder LM. Exp. 2006. The episodic buffer: a new component of working memory? Trends. Papagno C. Douglass S. ed. Richard Allen. Cogn.annualreviews. Pickering SJ. Qin Y. Psychol. J. Psychol. Bothell D. Working memory and the vividness of imagery. Hitch GJ. J. The Psychology of Memory. Ecob JR. as we begin So where does this leave our early question of to fill in the empty spaces on the theoretical what makes a good theory? Clearly. 2006. Oxford. Psychol. Byrne MD. Hitch GJ. Press Baddeley AD. Binding in visual working memory: the role of the episodic buffer. semantic and formal similarity. The influence of acoustic and semantic similarity on long-term memory for word sequences. Q. Gathercole SE. Psychol. Cambridge. pp. Oxford. Downloaded from www. Am. 1960. Psychol. 1988. Press Baddeley AD. by California State University . How does acoustic similarity influence short-term memory? Q. Exp. Exp. Exp. Working memory: activation limitations on retrieval. In The Psychology of Learning and Motivation: Advances in Research and Theory. MA: MIT Press Baars BJ. and Controversies 25 .org • Working Memory: Theories.

1974. Working memory. Mishkin M. Baddeley AD. Imagery and visual working memory. 27:586–95 Baddeley AD. 1953. Larsen JD. 55:627–52 Cermak LS. Sentence comprehension and phonological memory: some neu- ropsychological evidence. Hitch GJ. 1999. Vallar G. UK: Cambridge Univ. Psychol. Exploring the articulatory loop. Psychol. 509–29. ed. Hitch GJ. Memory for serial order: a network model of the phonological loop and its timing. Preece T. J. Thomson N. R Nickerson. ed. 1:141–50 Cermak LS. Cambridge. ed. London: Erlbaum Baddeley AD. 2009. Acoustic masking in primary memory. acoustic confusion and memory span. A Miyake. Grant W. Q. Press Baddeley AD. Lieberman K. London: Academic Baddeley AD. Verbal Learn. pp. Cognitive resources and the learning of human-computer dialogs. Verbal Behav. Lang. Recency re-examined. Psychol. Rev. Press Brooks LR. Depth of processing and retention of words by alcoholic Korsakoff patients.Channel Islands on 09/12/12. Q. The phonological loop unmasked? A comment on the evidence for a “perceptual-gestural” alternative. Papagno C. Psychol. London: Erlbaum Barnard PJ. 1999. In Attention and Performance VI. Press 26 Baddeley . Wilson BA. A revised model of short-term memory and long-term learning of verbal sequences. Mem. Interactive cognitive subsystems: a psycholinguistic approach to short-term memory. Welsh A. pp. In Models of Working Memory. J. A Miyake. NJ: Erlbaum Baddeley AD. NJ: Erlbaum by California State University . In Interfacing Thought: Cognitive Aspects of Human-Computer Interaction. Information. 1971. 1970. Hitch GJ. ed. 89:132–36 Baddeley AD. Lang. UK: Psychol. Oscillator-based memory for serial order. Neurosci. Cogn. Psychol. Verbal Behav. MRC Appl. UK Braithwaite RB. 47–89. Verbal Learn. 1985.: Gen. Unit. Verbal Learn. Wight E. J. 62–101. 1977. pp. A Ellis. Downloaded from www. 2001. Brain Lang. In Attention and Performance V. 1974. J. 1987. Exp. J. Psychol. Baddeley AD. Logie RH. 133:83–100 Bowman H. Working memory: the multiple component model. In Attention and Performance XII: The Psychology of Reading. Lang. In Models of Working Mem- Annu. 1980.63:1-29. Word length and the structure of short-term memory. S Dornic. Time constraints and resource sharing in adults’ working memory spans. 205–17. Learn. Vallar G. Working memory and binding in sentence recall. Hum. Psychol. 28–61. Some analyses of the verbal encoding deficit of alcoholic Korsakoff patients. Exp. Vallar G. J. ed. In Attention and Performance VIII. Mem. Rev. J. 2012. In The Psychology of Learning and Motivation: Advances in Research and Theory. Hillsdale. Cambridge. pp. Levy BA. M Coltheart. Exp.annualreviews. Bernardin S. 1988. 1999.. Hitch GJ. Buchanan M. Cambridge. 1975b. ed. New York: Academic Baddeley AD. J. J. 1987. 106:551–81 Burgess N. pp. Hitch GJ. 2000. Thomson N. pp. S Dornic. J. Psychol. 647–67. Camos V. Allen RJ. J. Warrington EK. 2005. Hull AJ. Hillsdale. When long-term learning depends on short-term storage. Butters N. J. Lewis V. Exp. 2004. ed. 60:497–504 Baddeley AD. Scientific Explanation. Psychol. Reale L. 55:429–32 Cowan N. JM Carroll. 2007. UK: Cambridge Univ. Press Cowan N. ory: Mechanisms of Active Maintenance and Executive Control. Amnesia and the distinction between long. For personal use only. P Shah. An embedded-processes model of working memory. Cambridge. 14:575–89 Baddeley AD. J. Br. 521–39. 15:17–32 Conrad R. 2011. UK: Cambridge Univ. Hove. 61:438–56 Baddeley AD. pp. Can. Spatial and verbal components in the act of recall. Rev. Moreines J. Exp. Working Memory Capacity. 4:165–74 Colle HA. Psychol. 107:127–81 Burgess N. 13:357–69 Baddeley AD. GA Bower. pp. 197–258. Verbal Behav. PMA Rabbitt. Cambridge. 1984. Vargha-Khadem F. 1964. Spatial working memory. In Progress in the Psychology of Language. Psychol. 1978. Is there a future in architectural modeling of mind? Paper presented at Philip Barnard retirement symp. 2006. 112–58. ed. pp. J. 1976. Hulme C. MA: MIT Press Barrouillet P.and short-term memory. A 36:233–52 Baddeley AD. 22:349–68 Brown GDA. ed. 9:176–89 Barnard PJ. Semantic coding and short-term memory. 1975a. Mem. P Shah. Preserved recognition in a case of developmental amnesia: implications for the acquisition of semantic memory.

UK Johnson MK.annualreviews. 2010. pp.Channel Islands on 09/12/12. Verbal Behav. Psychol. Cohen NJ. 87–104. The role of verbal output time and the effects of word-length on immediate memory. Irrelevant tones produce an irrelevant speech effect: implications for phono- logical coding in working memory. Working memory and language comprehension: a meta-analysis. Allamano N. Natl. even at short lags. 61:97–111 Hurlstone MJ. 23:83–94 Gathercole SE. Working Memory and Learning: A Practical Guide. Mem. Mishkin M. In Models of Working Memory: Mechanisms of Active Maintenance and Executive Control.or between- stream similarity? J. 102–34. Psychol. general fluid intelligence. L Weiskrantz. J Andrade. Objects. working memory capacity and two-factor theory of cognitive control. For personal use only. Psychol. Appl. Neural working memory. Psychol. Ambridge B. Mem. Mem. 40:177–90 Gathercole SE. Baddeley A. P Shah. Verbal Behav. Macken WJ. 2001. USA 98:8101–6 Emerson • Working Memory: Theories. UK: Psychol. 40:1–16 Glanzer M. J. Heinze HJ. 48:148–68 by California State University . 2009. 1993. Raye CL. London: Sage Gathercole SE. 1993. J. Storage mechanisms in recall. Carpenter PA. Mem. 1972. Hove. 1995. New York: Cambridge Univ. Awareness and Control. UK: Clarendon Jones DM. Psychol. Phonological similarity in the irrelevant speech effect: within. Lang. Dev. 11:671–84 Dale HCA. Levels of processing: a framework for memory research. streams and threads of auditory attention. 26:8352–59 Henson R.: Learn. Johnson T. Long-term working memory. Neurosci. Models. 1972. 2001. PhD dissert. Sci. Proc. New York: Elsevier Engle RW. Lang. J. J. Kane MJ. 151–74. Lockhart RS. Working memory skills and educational at- tainment: evidence from National Curriculum assessments at 7 and 14 years of age. J. Is nonword repetition a test of phonological memory or long-term knowledge? It all depends on the nonwords. Topography of cognition: parallel distributed networks in primate association cortex. Exp. Unpubl. ed. J. 19:369–81 Jones DM. 1999.. 2003. 145–99. Downloaded from www. Individual differences in working memory and reading. 2004a. Perceptual organization masquerading as phonological storage: further support for a perceptual-gestural view of short-term memory. MA: MIT Press Gathercole SE. J. Keller TA. Pickering SJ. Mem. 21:103–15 www. 2006. Second thoughts versus second looks: an age-related deficit in reflectively refreshing just-activated information. Knight C. Macken WJ. GH Bower. and functions of the prefrontal cortex. Exp. Cogn. Tuholski Engle RW. Merikle PM. B Ross. 1973. The role of inner speech in task switching: a dual-task investigation. pp. Cogn. pp. Macken WJ. Sci. 13:64–67 Jones D. Annu.annualreviews. Reeder JA. Flude B. 1992. Vargha-Khadem F. Stegmann Z. 1999. 1995. Flores L. pp. and Controversies 27 . 2008. Psychonom. 11:137–56 Hannula DE. Psychol. Psychol. Lang. Press Hitch GJ. Day L. Br. Psychol. 64:1–8 Daneman M. ed. ed. Lang. Verbal Learn. In The Psychology of Learning and Motivation: Advances in Research and Theory. Miyake A. Wearing H. Pickering SJ. J. York. Rev. Slave to the rhythm: experimental tests of a model for verbal short-term memory and long-term sequence learning. Neurosci. 129–93. 1995. AD Baddeley. 102:211–45 Fodor JA. The problem of serial order in visuospatial short-term memory. Mem. Mitchell KJ. Individual differences in working memory capacity and what they tell us about controlled attention. 3:422–33 Della Sala S. Alloway TP. A Miyake. Mem. Cogn. In Working Memory in Perspective. Hughes RW. 19:450–66 Daneman M. Executive attention.: Learn. 2006. J. New York: Academic Goldman-Rakic PW. Cambridge.63:1-29. ed. Tranel D. pp. Wilson L. ed. Acad. Rev. Brain activity evidence for recognition without recollection after early hippocampal damage. 54:265–81 Jones DM. Rev. Kane MJ. In The Psychology of Learning and Motivation. 1980. The long and the short of it: relational memory impairments in amnesia. Rev. Pattern span: a tool for unwelding visuo-spatial memory. 1983. 1988. In Attention: Selection. Gray C. The Modularity of Mind. Bull. Univ. Verbal Learn. Saults JS. 2004. Annu. Neuropsychologia 37:1189–99 Düzel E. Press Ericsson KA. Short-term memory for visual information. Oxford. 2002. Cowan N. 2004b. Cognit. The structure of working memory from 4 to 15 years of age. 1996. Kintsch W. 2012. 31:1–17 Craik FIM. Burgess N.

Exp. Norris D. Witzki AH. J. Br. Disruption of verbal STM by irrelevant speech. pp. Neurol. J. and Age. ed. Exp. 85:220–24 Lakatos I. 2005. Moore KS. 1970. Exp. UK: Oxford Univ. Wager TD. Johnson AL. Parkinson SR. 1990. Mem. The primacy model: a new model of immediate serial recall. Psychol. 1959. Kazén M. Q. J. Psychol. Cocchini G. 20:240–45 Logie RH. Howenter A. London: Hutchison Posner MI. Psychol. Nee DE. UK: Erlbaum Logie RH. Lustig CA. Symposium. 1–18. Baddeley AD. 51:45–100 Olson IR. 1967. J. Exp. Short-term memory while shadowing: recall of visually and aurally presented letters. 1997. The Logic of Scientific Discovery. 18:251–69 Nairne JS. Berman MG. Psychol. Sci. Exp. ed. Q. Neurosci. Sci. 1951. 38A:229–47 Logie RH. 2012. Rev. Annu. 59:193–224 Kaschel R. Cognit. Baddeley AD. Double dissociations in visual and spatial short-term memory. UK: Cambridge Univ. The directive function of speech in development and dissolution. Moore KS. 1959. Nature 390:279–81 Luria AR. 133:355–81 Kroll NE. Cogn. 1949. In Cerebral Mechanisms in Behavior: The Hixon by California State University . Is there a specific capacity for dual task co-ordination? Evidence from Alzheimer’s disease. 2010. Remembering over the short-term: the case against the standard model. Percept. Downloaded from www. 2000. A 56:1249–68 Larsen JD. Proofs and Reputations. Shallice T. 27:272–77 Miyake A. 1998. Alzheimer’s disease. A feature model of immediate memory. Baddeley A. Della Sala S. 56:132–43 Page MPA. 1995. Bull. Cambridge. 18:1087–97 Osgood CE. 1971. Psychol. In Consciousness and Self-Regulation. 2000. The unity and diversity of executive functions and their contributions to complex “frontal lobe” tasks: a latent variable analysis. Stark M.annualreviews. Annu. Baddeley AD. 1976. The irrelevant sound effect: what needs modeling. Neuropsychology 18:504–13 Logie RH. Psychol. 1986. articulatory suppression and manual tapping: Do they have a common source? Q. Rev. The similarity paradox in human learning: a resolution. Med. L Phillips. Bieber SL. Rev. J. Visual working memory is impaired when the medial temporal lobe is damaged. P McLeod. 2002. Parks T. but not ageing or depression. Lewis RL. Learn. Emerson MJ. 2004. Word 15:341–52 Milner B. 161–90. Psy- chophys. Friedman NP. 16:283–90 Phillips WA. Visuo-spatial processing in working memory. 2003. LA Jeffress. 53:53–81 Norman DA. 2009. Andrade J. J. 1971. 105:761– 81 Page MPA. Phonological similarity and the irrelevant speech effect: implications for models of short-term verbal memory. Norris DG. Psychol. For personal use only. 2004. Visual similarity effects in immediate serial recall. The problem of serial order in behavior. The mind and brain of short-term memory. Baddeley AD. Visuo-Spatial Working Memory. Memory 8:145–57 Lashley KS. J Duncan. Keele SW. On the distinction between sensory storage and short-term visual memory. GE Schwartz. The chronometrics of task-set control. Wynn V. pp. Psychol. Psychol. Della Sala S. Control. D Shapiro. J. pp. Press Larsen J. 1974. 2008. New York: John Wiley Annu. Psychol. Della Sala S.: Gen. ed. Chatterjee A. 22:73–4 Popper K. Attention to action: willed and automatic control of behaviour. Cogn. Psychol. Psychon. Interhemispheric differences in the localization of psychological processes in man. 1986. Press Nairne JS. and a tentative model. affects dual-tasking. Motiv. Psychol. Oxford. Curr. Rev. Design for a working memory. 2000. 112–36. Decay of visual information from a single letter. 41:49–100 Monsell S. 2006. Psychol.Channel Islands on 09/12/12. New York: Plenum Oberauer K. Science 158:137–39 28 Baddeley . Vogel EK. Advances in Research and Theory. 2003. 53A:626–46 Luck SJ. Psychol. Logie RH. RJ Davidson. Exp. part I. Q. Hove. The functional organisation and the capacity limits of working memory. 56A:1289–300 Phillips WA. J. In Measuring the Mind: Speed.63:1-29. 256:1860–68 Klauer KC. Rev. Zhao Logie RH. Dir. 2011. Reaction time and short-term visual memory. Jonides J. The capacity of visual working memory for features and conjunctions.

ed. 28:341–48 Unsworth N. Rev. Psychol. 51:5–16 www. Psychol. Perfect TJ. UK: Cambridge Univ. Percept. 26:709–77 Saito S. Hitch GJ. Squire LR. 1997. Press by California State University . 22:261–73 Shrager Y. 2001. Organ. 2007. On the interpretation of working memory span in adults. 261–77. For personal use only. Jonides J.Channel Islands on 09/12/12. New York: Oxford Univ.: Gen. 2002. Posner MI. et al. 114:104– 32 Vogel EK. 28:4818–22 Smith EE. Rev. Learn. Cameron K. 2004. pp. 2004. Hove. 2004. Press Shallice T. UK: Erlbaum Toulmin S. 20:165–82 Ruchkin DS. Short-term retention of visual and kinesthetic information. Baddeley AD. Exp. Brain Sci. Rev. 2003. Independent functioning of verbal memory stores: a neuropsychological study. Norman DA. Cognit. RT Knight. In Principles of Frontal Lobe Function. Neural correlates of working memory for sign language. Anthropol. Space and movement in working memory. Neurosci. Mem. Woodman GF. Binding in short-term visual memory. Neurobiol. Treisman AM. Hum. Cognit. Working memory and the organization of brain systems. Cogn. Bradley A. 1988. Exp. Grafman J. Rudner M. Hitch G. Fearneyhough C. J. Cognit. Cogn. 24:83–93 Rönnberg J. and objects in visual working memory. 1966. 1993. Soc. J. 27:92–114 Waugh NC. 1986. The Philosophy of Science. Eur. Affect. Perform. 131:48–64 Williamson V. Curr. Psychol. Musicians’ and nonmusicians’ short-term memory for verbal and musical sequences: comparing phonological similarity and pitch proximity. 16:863–91 Robbins T. Working Memory. 1953. DT Stuss. Psychonom. Hopkins RO. Psychol. The effects of age and task domain on dual task performance: a meta-analysis. 2010. J. 1970. Konick AF. Exp. Working memory in chess. 1990. 72:89–104 Wheeler ME. Mem. Berndt RS. Baddeley A. Coolidge FL. 2000.annualreviews. Barnard PJ. On the nature of forgetting and the processing-storage relationship in reading span Shallice T. Cognit. Hutton U. 42A:291–304 Squire LR. Q. 2008. London: Hutchison Towse JN. Exp. J. Mem. Cambridge. Henderson E.63:1-29. Psychol. Miyake A. Working memory retention systems: a state of activated long-term memory. Luck SJ. From Neuropsychology to Mental Structure. J. Annu. Q. 2004. 1965. Cognition and Change: Remodeling Depressive Thought. conjunctions. Psychol. Beyond symbolism and language: an introduction to Supplement 1. Warrington EK. • Working Memory: Theories. Bull. 24:263–65 Shallice T.annualreviews. J. 1:71–86 Riby LM. and Controversies 29 . Levy DA. Stollery B. Engle RW. 50:425–43 Salame P. 38:163–75 Wynn T. 1996. Downloaded from www. Storage of features. Models. Pendleton LR. 2012. Fractionation of the supervisory system. Mem. 2002. Mem. Psychol. Lang. 33:5–42 Smyth MM. Working memory: a view from neuroimaging. Behav. Behav. Psychol. Barker D. The nature of individual differences in working memory capacity: active maintenance in primary memory and controlled search from secondary memory. Primary memory. Psychol. Ingvar M. 82:171–77 Teasdale JD. Memory systems of the brain: a brief history and current perspective. Perform. 2010.: Hum. Brain Res. Phonological factors in STM: similarity and the unattended speech effect.

Shaun Nichols.63:1-29. Rauschecker. Seyfarth and Dorothy L. Wandell. Developmental Psychobiology Learning to See Words Brian A. Rev. Andreas M. and Jason D. Downloaded from www.annualreviews. Cheney p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 153 Emotional. Wagner p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 101 Neuroscience of Learning Fear Extinction as a Model for Translational Neuroscience: Ten Years of Progress Mohammed R. 2012 Contents Prefatory Working Memory: Theories. Philip Robbins. Yeatman p p p p p p p p p p p p p p p p p p p p p31 Memory Remembering in Conversations: The Social Sharing and Reshaping of Memories William Hirst and Gerald Echterhoff p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p55 Judgment and Decision Making Experimental Philosophy Joshua Knobe. Quirk p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 129 Comparative Psychology The Evolutionary Origins of Friendship Robert M. Morality. Evolution Paul Bloom p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 179 vi . and Personality Development Religion. and Tamler Sommers p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p81 Brain Imaging/Cognitive Neuroscience Distributed Representations in Memory: Insights from Functional Brain Imaging Jesse Rissman and Anthony D. Wesley Buckwalter. Annual Review of Psychology Volume 63. Psychol. Models.Channel Islands on 09/12/12. For personal use only. and Controversies Alan Baddeley p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 1 by California State University . Milad and Gregory J. Annu. Social. Hagop Sarkissian.

Psychol. Niedenthal and Markus Brauer p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 259 by California State University . Levine p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 397 Education of Special Populations Rapid Automatized Naming (RAN) and Reading Fluency: Implications for Understanding and Treatment of Reading Disabilities Elizabeth S. Social Personality. Sanchez and Edward L. Frith and Uta Frith p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 287 Personality Processes Personality Processes: Mechanisms by Which Personality Traits “Get Outside the Skin” Sarah E. Kammeyer-Mueller p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 341 The Individual Experience of Unemployment Connie R. 2012. Downloaded from www. Deary p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 453 Research Methodology Decoding Patterns of Human Brain Activity Frank Tong and Michael S. Hampson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 315 Work Attitudes Job Attitudes Timothy A. Pratte p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 483 Contents vii .annualreviews. Judge and John D. Wanberg p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 369 Job/Work Analysis The Rise and Fall of Job Analysis and the Future of Work Analysis Juan I. Social Motivation. For personal use only. Social Emotion Social Functionality of Human Emotion Paula M. and Terrorism: Pathways of Risk and Resilience Ann S. Annu. War. Masten and Angela J.63:1-29. Social Neuroscience Mechanisms of Social Cognition Chris D. Norton and Maryanne Wolf p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 427 Human Abilities Intelligence Ian J.Channel Islands on 09/12/12. Narayan p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 227 Social Development. Adulthood and Aging Consequences of Age-Related Cognitive Declines Timothy Salthouse p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 201 Development in Societal Context Child Development in the Context of Disaster.

Volumes 53–63 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 598 Errata An online log of corrections to Annual Review of Psychology articles may be found at http://psych. Podsakoff p p p p p p p p p p p p p p p p p p p p 539 Neuroscience Methods Neuroethics: The Ethical. Volumes 53–63 p p p p p p p p p p p p p p p p p p p p p p p p p p p 593 Annu. Psychol. Podsakoff. Human Intracranial Recordings and Cognitive Neuroscience Roy Mukamel and Itzhak Fried p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 511 Sources of Method Bias in Social Science Research and Recommendations on How to Control It Philip M. and Nathan P. MacKenzie. For personal use only. 2012.Channel Islands on 09/12/12. Rev. Downloaded from www.AnnualReviews. Legal. Farah p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 571 Indexes by California State University .org/errata. Cumulative Index of Contributing Cumulative Index of Chapter Titles.shtml viii Contents .63:1-29. Scott B.annualreviews. and Societal Impact of Neuroscience Martha J.