Editing Working memory (section)

== Capacity ==
Working memory is widely acknowledged as having limited capacity. An early quantification of the capacity limit associated with short-term memory was the "[[The Magical Number Seven, Plus or Minus Two|magical number seven]]" suggested by Miller in 1956.<ref name="miller">{{cite journal | vauthors = Miller GA | title = The magical number seven plus or minus two: some limits on our capacity for processing information | journal = Psychological Review | volume = 63 | issue = 2 | pages = 81–97 | date = March 1956 | pmid = 13310704 | doi = 10.1037/h0043158 | s2cid = 15654531 }} Republished: {{cite journal | vauthors = Miller GA | title = The magical number seven, plus or minus two: some limits on our capacity for processing information. 1956 | journal = Psychological Review | volume = 101 | issue = 2 | pages = 343–352 | date = April 1994 | pmid = 8022966 | doi = 10.1037/0033-295X.101.2.343 | hdl-access = free | hdl = 11858/00-001M-0000-002C-4646-B }}</ref> Miller claimed that the information-processing capacity of young adults is around seven elements, referred to as "chunks", regardless of whether the elements are digits, letters, words, or other units. Later research revealed this number depends on the category of chunks used (e.g., span may be around seven for digits, six for letters, and five for words), and even on features of the [[chunking (psychology)|chunks]] within a category. For instance, attention span is lower for longer words than short words. In general, memory span for verbal contents (digits, letters, words, etc.) depends on the phonological complexity of the content (i.e., the number of phonemes, the number of syllables),<ref>{{Cite journal| vauthors = Service E |date=1998-05-01|title=The Effect of Word Length on Immediate Serial Recall Depends on Phonological Complexity, Not Articulatory Duration|journal=The Quarterly Journal of Experimental Psychology Section A|volume=51|issue=2|pages=283–304|doi=10.1080/713755759|s2cid=220062579|issn=0272-4987}}</ref> and on the lexical status of the contents (whether the contents are words known to the person or not).<ref>{{Cite journal| vauthors = Hulme C, Roodenrys S, Brown G, Mercer R |date=November 1995 |title=The role of long-term memory mechanisms in memory span |journal=British Journal of Psychology |volume=86 |issue=4 |pages=527–36 |doi=10.1111/j.2044-8295.1995.tb02570.x}}</ref> Several other factors affect a person's measured span, and therefore it is difficult to pin down the capacity of short-term or working memory to a number of chunks. Nonetheless, Cowan proposed that working memory has a capacity of about four chunks in young adults (and fewer in children and old adults).<ref>{{cite journal | vauthors = Cowan N | title = The magical number 4 in short-term memory: a reconsideration of mental storage capacity | journal = The Behavioral and Brain Sciences | volume = 24 | issue = 1 | pages = 87–185 | date = February 2001 | pmid = 11515286 | doi = 10.1017/S0140525X01003922 | doi-access = free }}</ref>

In the visual domain, some investigations report no fixed capacity limit with respect to the total number of items that can be held in working memory. Instead, the results argue for a limited resource that can be flexibly shared between items retained in memory (see below in Resource theories), with some items in the focus of attention being allocated more resource and recalled with greater precision.<ref name="Changing concepts of working memory">{{cite journal | vauthors = Ma WJ, Husain M, Bays PM | title = Changing concepts of working memory | journal = Nature Neuroscience | volume = 17 | issue = 3 | pages = 347–356 | date = March 2014 | pmid = 24569831 | pmc = 4159388 | doi = 10.1038/nn.3655 }}</ref><ref name="The precision of visual working mem">{{cite journal | vauthors = Bays PM, Catalao RF, Husain M | title = The precision of visual working memory is set by allocation of a shared resource | journal = Journal of Vision | volume = 9 | issue = 10 | pages = 7.1–711 | date = September 2009 | pmid = 19810788 | pmc = 3118422 | doi = 10.1167/9.10.7 }}</ref><ref name="Temporal dynamics of encoding, stor">{{cite journal | vauthors = Bays PM, Gorgoraptis N, Wee N, Marshall L, Husain M | title = Temporal dynamics of encoding, storage, and reallocation of visual working memory | journal = Journal of Vision | volume = 11 | issue = 10 | pages = 6 | date = September 2011 | pmid = 21911739 | pmc = 3401684 | doi = 10.1167/11.10.6 }}</ref><ref name="A review of visual memory capacity">{{cite journal | vauthors = Brady TF, Konkle T, Alvarez GA | title = A review of visual memory capacity: Beyond individual items and toward structured representations | journal = Journal of Vision | volume = 11 | issue = 5 | pages = 4 | date = May 2011 | pmid = 21617025 | pmc = 3405498 | doi = 10.1167/11.5.4 }}</ref>

Whereas most adults can repeat about seven digits in correct order, some individuals have shown impressive enlargements of their digit span—up to 80 digits. This feat is possible by extensive training on an encoding strategy by which the digits in a list are grouped (usually in groups of three to five) and these groups are encoded as a single unit (a chunk). For this to succeed, participants must be able to recognize the groups as some known string of digits. One person studied by Ericsson and his colleagues, for example, used an extensive knowledge of racing times from the history of sports in the process of coding chunks: several such chunks could then be combined into a higher-order chunk, forming a hierarchy of chunks. In this way, only some chunks at the highest level of the hierarchy must be retained in working memory, and for retrieval the chunks are unpacked. That is, the chunks in working memory act as retrieval cues that point to the digits they contain. Practicing memory skills such as these does not expand working memory capacity proper: it is the capacity to transfer (and retrieve) information from long-term memory that is improved, according to Ericsson and Kintsch (1995; see also Gobet & Simon, 2000<ref name="Gobet F 2000 551–70">{{cite journal | vauthors = Gobet F | title = Some shortcomings of long-term working memory | journal = British Journal of Psychology | volume = 91 | issue = Pt 4 | pages = 551–570 | date = November 2000 | pmid = 11104178 | doi = 10.1348/000712600161989 | url = http://bura.brunel.ac.uk/handle/2438/807 | type = Submitted manuscript }}</ref>).

=== Measures and correlates ===
Working memory capacity can be tested by a variety of tasks. A commonly used measure is a dual-task paradigm, combining a [[memory span]] measure with a concurrent processing task, sometimes referred to as "complex span". Daneman and Carpenter invented the first version of this kind of task, the "[[reading span]]", in 1980.<ref>{{Cite journal| vauthors = Daneman M, Carpenter PA |date=August 1980 |title=Individual differences in working memory and reading |journal=Journal of Verbal Learning & Verbal Behavior |volume=19 |issue=4 |pages=450–66 |doi=10.1016/S0022-5371(80)90312-6|s2cid=144899071 }}</ref> Subjects read a number of sentences (usually between two and six) and tried to remember the last word of each sentence. At the end of the list of sentences, they repeated back the words in their correct order. Other tasks that do not have this dual-task nature have also been shown to be good measures of working memory capacity.<ref>{{Cite journal| vauthors = Oberauer K, Süß HM, Schulze R, Wilhelm O, Wittmann WW |date=December 2000|title=Working memory capacity—facets of a cognitive ability construct|journal=Personality and Individual Differences|volume=29|issue=6|pages=1017–45|doi=10.1016/S0191-8869(99)00251-2 |s2cid=143866158 }}</ref> Whereas Daneman and Carpenter believed that the combination of "storage" (maintenance) and processing is needed to measure working memory capacity, we know now that the capacity of working memory can be measured with short-term memory tasks that have no additional processing component.<ref>{{cite journal | vauthors = Unsworth N, Engle RW | title = On the division of short-term and working memory: an examination of simple and complex span and their relation to higher order abilities | journal = Psychological Bulletin | volume = 133 | issue = 6 | pages = 1038–1066 | date = November 2007 | pmid = 17967093 | doi = 10.1037/0033-2909.133.6.1038 }}</ref><ref>{{Cite journal| vauthors = Colom R, Abad FJ, Quiroga MÁ, Shih PC, Flores-Mendoza C |year=2008|title=Working memory and intelligence are highly related constructs, but why?|journal=Intelligence|volume=36|issue=6|pages=584–606|doi=10.1016/j.intell.2008.01.002}}</ref> Conversely, working memory capacity can also be measured with certain processing tasks that do not involve maintenance of information.<ref>{{Cite journal |vauthors = Oberauer K, Süß HM, Wilhelm O, Wittman WW |year=2003 |title=The multiple faces of working memory – storage, processing, supervision, and coordination |doi=10.1016/s0160-2896(02)00115-0 |journal=Intelligence |volume=31 |issue=2 |pages=167–193 |s2cid=14083639 |url=https://www.zora.uzh.ch/id/eprint/97155/1/intelligence.pdf}}</ref><ref>{{cite journal | vauthors = Chuderski A | title = The relational integration task explains fluid reasoning above and beyond other working memory tasks | journal = Memory & Cognition | volume = 42 | issue = 3 | pages = 448–463 | date = April 2014 | pmid = 24222318 | pmc = 3969517 | doi = 10.3758/s13421-013-0366-x }}</ref> The question of what features a task must have to qualify as a good measure of working memory capacity is a topic of ongoing research.

Recently, several studies of visual working memory have used delayed response tasks. These use analogue responses in a continuous space, rather than a binary (correct/incorrect) recall method, as often used in visual change detection tasks. Instead of asking participants to report whether a change occurred between the memory and probe array, delayed reproduction tasks require them to reproduce the precise quality of a visual feature, e.g. an object's location, orientation or colour.<ref name="Changing concepts of working memory"/><ref name="The precision of visual working mem"/><ref name="Temporal dynamics of encoding, stor"/><ref name="A review of visual memory capacity"/> In addition, the combination of visual perception such as within objects and colors can be used to improve memory strategy through elaboration, thus creating reinforcement within the capacity of working memory.<ref>{{cite journal |last1=Sobrinho |first1=Nuno D. |last2=Souza |first2=Alessandra S. |title=The interplay of long-term memory and working memory: When does object-color prior knowledge affect color visual working memory? |journal=Journal of Experimental Psychology: Human Perception and Performance |date=February 2023 |volume=49 |issue=2 |pages=236–262 |doi=10.1037/xhp0001071 |pmid=36480376 |hdl=10216/147912 |url=https://psyarxiv.com/8a2jw/ |hdl-access=free }}</ref>

Measures of working-memory capacity are strongly related to performance in other complex cognitive tasks, such as reading comprehension, problem solving, and with measures of [[intelligence quotient]].<ref>{{cite journal | vauthors = Conway AR, Kane MJ, Engle RW | title = Working memory capacity and its relation to general intelligence | journal = Trends in Cognitive Sciences | volume = 7 | issue = 12 | pages = 547–552 | date = December 2003 | pmid = 14643371 | doi = 10.1016/j.tics.2003.10.005 | s2cid = 9943197 }}</ref>

Some researchers have argued<ref>{{cite journal | vauthors = Engle RW, Tuholski SW, Laughlin JE, Conway AR | title = Working memory, short-term memory, and general fluid intelligence: a latent-variable approach | journal = Journal of Experimental Psychology. General | volume = 128 | issue = 3 | pages = 309–331 | date = September 1999 | pmid = 10513398 | doi = 10.1037/0096-3445.128.3.309 | s2cid = 1981845 }}</ref> that working-memory capacity reflects the efficiency of executive functions, most notably the ability to maintain multiple task-relevant representations in the face of distracting irrelevant information; and that such tasks seem to reflect individual differences in the ability to focus and maintain attention, particularly when other events are serving to capture attention. Both working memory and executive functions rely strongly, though not exclusively, on frontal brain areas.<ref name="Kane MJ, Engle RW 2002 637–71">{{cite journal | vauthors = Kane MJ, Engle RW | title = The role of prefrontal cortex in working-memory capacity, executive attention, and general fluid intelligence: an individual-differences perspective | journal = Psychonomic Bulletin & Review | volume = 9 | issue = 4 | pages = 637–671 | date = December 2002 | pmid = 12613671 | doi = 10.3758/BF03196323 | doi-access = free }}</ref>

Other researchers have argued that the capacity of working memory is better characterized as the ability to mentally form relations between elements, or to grasp relations in given information. This idea has been advanced, among others, by Graeme Halford, who illustrated it by our limited ability to understand statistical interactions between variables.<ref>{{cite journal | vauthors = Halford GS, Baker R, McCredden JE, Bain JD | title = How many variables can humans process? | journal = Psychological Science | volume = 16 | issue = 1 | pages = 70–76 | date = January 2005 | pmid = 15660854 | doi = 10.1111/j.0956-7976.2005.00782.x | s2cid = 9790149 }}</ref> These authors asked people to compare written statements about the relations between several variables to graphs illustrating the same or a different relation, as in the following sentence: "If the cake is from France, then it has more sugar if it is made with chocolate than if it is made with cream, but if the cake is from Italy, then it has more sugar if it is made with cream than if it is made of chocolate". This statement describes a relation between three variables (country, ingredient, and amount of sugar), which is the maximum most individuals can understand. The capacity limit apparent here is obviously not a memory limit (all relevant information can be seen continuously) but a limit to how many relationships are discerned simultaneously.{{citation needed|date=July 2022}}

=== Experimental studies of working-memory capacity ===
There are several hypotheses about the nature of the capacity limit. One is that a limited pool of cognitive resources is needed to keep representations active and thereby available for processing, and for carrying out processes.<ref name=":0">{{cite journal | vauthors = Just MA, Carpenter PA | title = A capacity theory of comprehension: individual differences in working memory | journal = Psychological Review | volume = 99 | issue = 1 | pages = 122–149 | date = January 1992 | pmid = 1546114 | doi = 10.1037/0033-295X.99.1.122 | s2cid = 2241367 }}</ref> Another hypothesis is that memory traces in working memory decay within a few seconds, unless refreshed through rehearsal, and because the speed of rehearsal is limited, we can maintain only a limited amount of information.<ref>{{cite journal | vauthors = Towse JN, Hitch GJ, Hutton U | title = On the interpretation of working memory span in adults | journal = Memory & Cognition | volume = 28 | issue = 3 | pages = 341–348 | date = April 2000 | pmid = 10881551 | doi = 10.3758/BF03198549 | doi-access = free }}</ref> Yet another idea is that representations held in working memory interfere with each other.<ref>{{cite journal | vauthors = Waugh NC, Norman DA | title = PRIMARY MEMORY | journal = Psychological Review | volume = 72 | issue = 2 | pages = 89–104 | date = March 1965 | pmid = 14282677 | doi = 10.1037/h0021797 }}</ref>

====Decay theories====

The assumption that the contents of short-term or working memory [[decay theory|decay]] over time, unless decay is prevented by rehearsal, goes back to the early days of experimental research on short-term memory.<ref>{{Cite journal| vauthors = Brown J |year=1958|title=Some tests of the decay theory of immediate memory|journal=Quarterly Journal of Experimental Psychology|volume=10|pages=12–21|doi=10.1080/17470215808416249|s2cid=144071312}}</ref><ref>{{cite journal | vauthors = Peterson LR, Peterson MJ | title = Short-term retention of individual verbal items | journal = Journal of Experimental Psychology | volume = 58 | issue = 3 | pages = 193–198 | date = September 1959 | pmid = 14432252 | doi = 10.1037/h0049234 }}</ref> It is also an important assumption in the multi-component theory of working memory.<ref>{{Cite book|title=Working memory| vauthors = Baddeley AD |publisher=Clarendon | volume = 11 |year=1986|location=Oxford | isbn = 978-0-19-852116-7 }}</ref> The most elaborate decay-based theory of working memory to date is the "time-based resource sharing model".<ref>{{cite journal | vauthors = Barrouillet P, Bernardin S, Camos V | title = Time constraints and resource sharing in adults' working memory spans | journal = Journal of Experimental Psychology. General | volume = 133 | issue = 1 | pages = 83–100 | date = March 2004 | pmid = 14979753 | doi = 10.1037/0096-3445.133.1.83 | s2cid = 604840 }}</ref> This theory assumes that representations in working memory decay unless they are refreshed. Refreshing them requires an attentional mechanism that is also needed for any concurrent processing task. When there are small time intervals in which the processing task does not require attention, this time can be used to refresh memory traces. The theory therefore predicts that the amount of forgetting depends on the temporal density (rate and duration) of attentional demands of the processing task—this density is called ''[[cognitive load]]''. The cognitive load depends on two variables, the rate at which the processing task requires individual steps to be carried out, and the duration of each step. For example, if the processing task consists of adding digits, then having to add another digit every half-second places a higher cognitive load on the system than having to add another digit every two seconds. In a series of experiments, Barrouillet and colleagues have shown that memory for lists of letters depends neither on the number of processing steps nor the total time of processing but on cognitive load.<ref>{{cite journal |last1=Barrouillet |first1=Pierre |last2=Bernardin |first2=Sophie |last3=Portrat |first3=Sophie |last4=Vergauwe |first4=Evie |last5=Camos |first5=Valérie |title=Time and cognitive load in working memory. |journal=Journal of Experimental Psychology: Learning, Memory, and Cognition |date=2007 |volume=33 |issue=3 |pages=570–585 |doi=10.1037/0278-7393.33.3.570 |pmid=17470006 |url=https://archive-ouverte.unige.ch/unige:88299 }}</ref>

====Resource theories====

Resource theories assume that the capacity of working memory is a limited resource that must be shared between all representations that need to be maintained in working memory simultaneously.<ref name="Changing concepts of working memory"/> Some resource theorists also assume that maintenance and concurrent processing share the same resource;<ref name=":0" /> this can explain why maintenance is typically impaired by a concurrent processing demand. Resource theories have been very successful in explaining data from tests of working memory for simple visual features, such as colors or orientations of bars. An ongoing debate is whether the resource is a continuous quantity that can be subdivided among any number of items in working memory, or whether it consists of a small number of discrete "slots", each of which can be assigned to one memory item, so that only a limited number of about 3 items can be maintained in working memory at all.<ref>{{cite journal | vauthors = van den Berg R, Awh E, Ma WJ | title = Factorial comparison of working memory models | journal = Psychological Review | volume = 121 | issue = 1 | pages = 124–149 | date = January 2014 | pmid = 24490791 | pmc = 4159389 | doi = 10.1037/a0035234 }}</ref>

====Interference theories====

Several forms of [[Interference theory|interference]] have been discussed by theorists. One of the oldest ideas is that new items simply replace older ones in working memory. Another form of interference is retrieval competition. For example, when the task is to remember a list of 7 words in their order, we need to start recall with the first word. While trying to retrieve the first word, the second word, which is represented in proximity, is accidentally retrieved as well, and the two compete for being recalled. Errors in serial recall tasks are often confusions of neighboring items on a memory list (so-called transpositions), showing that retrieval competition plays a role in limiting our ability to recall lists in order, and probably also in other working memory tasks. A third form of interference is the distortion of representations by superposition: When multiple representations are added on top of each other, each of them is blurred by the presence of all the others.<ref>{{cite journal |last1=Oberauer |first1=Klaus |last2=Lewandowsky |first2=Stephan |last3=Farrell |first3=Simon |last4=Jarrold |first4=Christopher |last5=Greaves |first5=Martin |title=Modeling working memory: An interference model of complex span |journal=Psychonomic Bulletin & Review |date=October 2012 |volume=19 |issue=5 |pages=779–819 |doi=10.3758/s13423-012-0272-4 |pmid=22715024 |url=https://www.zora.uzh.ch/id/eprint/63536/1/ZORA_NL_63536.pdf }}</ref> A fourth form of interference assumed by some authors is feature overwriting.<ref>{{Cite journal|doi=10.1016/j.jml.2006.08.009 |title=A formal model of capacity limits in working memory |date=November 2006 | vauthors = Oberauer K, Kliegl R |journal=Journal of Memory and Language |volume=55 |issue=4 |pages=601–26|doi-access=free }}</ref><ref>{{cite journal | vauthors = Bancroft T, Servos P | title = Distractor frequency influences performance in vibrotactile working memory | journal = Experimental Brain Research | volume = 208 | issue = 4 | pages = 529–532 | date = February 2011 | pmid = 21132280 | doi = 10.1007/s00221-010-2501-2 | s2cid = 19743442 }}</ref> The idea is that each word, digit, or other item in working memory is represented as a bundle of features, and when two items share some features, one of them steals the features from the other. As more items are held in working memory, whose features begin to overlap, the more each of them will be degraded by the loss of some features.{{citation needed|date=July 2022}}

==== Limitations ====
None of these hypotheses can explain the experimental data entirely. The resource hypothesis, for example, was meant to explain the trade-off between maintenance and processing: The more information must be maintained in working memory, the slower and more error prone concurrent processes become, and with a higher demand on concurrent processing memory suffers. This trade-off has been investigated by tasks like the reading-span task described above. It has been found that the amount of trade-off depends on the similarity of the information to be remembered and the information to be processed. For example, remembering numbers while processing spatial information, or remembering spatial information while processing numbers, impair each other much less than when material of the same kind must be remembered and processed.<ref>{{Cite journal|doi=10.1016/j.jml.2006.07.009 |title=The relationship between processing and storage in working memory span: Not two sides of the same coin |date=February 2007 | vauthors = Maehara Y, Saito S |journal=Journal of Memory and Language |volume=56 |issue=2 |pages=212–228}}</ref> Also, remembering words and processing digits, or remembering digits and processing words, is easier than remembering and processing materials of the same category.<ref>{{Cite journal|doi=10.1076/anec.6.2.99.784 |title=Selection from Working Memory: on the Relationship between Processing and Storage Components |date=June 1999 | vauthors = Li KZ |journal=Aging, Neuropsychology, and Cognition |volume=6 |issue=2 |pages=99–116}}</ref> These findings are also difficult to explain for the decay hypothesis, because decay of memory representations should depend only on how long the processing task delays rehearsal or recall, not on the content of the processing task. A further problem for the decay hypothesis comes from experiments in which the recall of a list of letters was delayed, either by instructing participants to recall at a slower pace, or by instructing them to say an irrelevant word once or three times in between recall of each letter. Delaying recall had virtually no effect on recall accuracy.<ref>{{cite journal |last1=Lewandowsky |first1=Stephan |last2=Duncan |first2=Matthew |last3=Brown |first3=Gordon D. A. |title=Time does not cause forgetting in short-term serial recall |journal=Psychonomic Bulletin & Review |date=October 2004 |volume=11 |issue=5 |pages=771–790 |doi=10.3758/bf03196705 |pmid=15732687 }}</ref><ref>{{cite journal |last1=Oberauer |first1=Klaus |last2=Lewandowsky |first2=Stephan |title=Forgetting in immediate serial recall: Decay, temporal distinctiveness, or interference? |journal=Psychological Review |date=2008 |volume=115 |issue=3 |pages=544–576 |doi=10.1037/0033-295X.115.3.544 |pmid=18729591 |url=https://api.research-repository.uwa.edu.au/ws/files/1546099/11204_PID11204.pdf }}</ref> The [[interference theory]] seems to fare best with explaining why the similarity between memory contents and the contents of concurrent processing tasks affects how much they impair each other. More similar materials are more likely to be confused, leading to retrieval competition.