What are the differences between long-term, short-term, and working memory?

What are the differences between long-term, short-term, and working memory? Cowan

Tag layer of the DHSS Meta-Tagging Project. The article text is reproduced from the staged copy named in sourceDesc and is not altered; the tags are held separately in standOff so that overlapping and nested spans are preserved.

What are the differences between long-term, short-term, and working memory? 2008 Progress in brain research 10.1016/s0079-6123(07)00020-9 Cowan2008-what-are-the-differences-between-long-term-s Cowan2008-what-are-the-differences-between-long-term-s.txt, the project's staged copy f6b42b5572ca87f2593dffa2a0e02e41f9cd2104 not-redistributable cognitive-psychology

the paper's own words

the tag layer; a sentence written by the tagger about the paper, always anchored to a quotation but not itself a quotation

0 spans located by searching the stored text for the quote's exact words, allowing any whitespace between them the three terms long-term, short-term and working memory, and whether the distinctions drawn between them are substantive or terminological Quotation withheld because this record is not cleared for redistribution. do short-term and long-term memory differ in kind — in temporal decay and chunk capacity — and is the further distinction between short-term and working memory a real one Quotation withheld because this record is not cleared for redistribution. separates the substantive question (is there decay, is there a capacity limit) from the terminological one (what shall we call the attention-related part), and concludes the second is a matter of usage while the first is not Quotation withheld because this record is not cleared for redistribution. existence Quotation withheld because this record is not cleared for redistribution. short-term, long-term and working memory distinctions Quotation withheld because this record is not cleared for redistribution. review review Quotation withheld because this record is not cleared for redistribution. mechanisms Short-term memory arises as a temporarily activated subset of long-term memory, part of which falls within a capacity-limited focus of attention Quotation withheld because this record is not cleared for redistribution. mechanisms Attention freed by an undemanding distractor task is used for an attention-based rehearsal that refreshes items, so recall falls as the distractor rate consumes that attention Quotation withheld because this record is not cleared for redistribution. mechanisms Working memory tasks correlate with aptitude because preventing covert rehearsal forces reliance on attention-demanding storage and processing Quotation withheld because this record is not cleared for redistribution. mechanisms A post-perceptual mask overwrites sensory-specific features in activated memory, leaving only categorical information inside the attention-protected focus Quotation withheld because this record is not cleared for redistribution. the whole argument is organised as contamination control: for each claimed property of short-term memory, name the confounds that could produce the same result, then ask which studies defeated them Quotation withheld because this record is not cleared for redistribution. a nuisance parameter is varied by four-fold across two modalities and the limit does not move, which converts an ambiguous number into a capacity rather than a rate Quotation withheld because this record is not cleared for redistribution. a cross-modal additivity prediction is used as the test: if one store holds everything, two modalities together cannot exceed the larger single modality alone Quotation withheld because this record is not cleared for redistribution. the confound is designed out of the experiment rather than corrected for: one trial per participant, gathered by putting the laboratory in a shopping mall Quotation withheld because this record is not cleared for redistribution. chunking is neutralised by choosing tasks in which grouping is impossible — an unpredictable list end, or a four-per-second rate — rather than by trying to measure the chunks Quotation withheld because this record is not cleared for redistribution. the chunk is made observable by defining it operationally through the participant's own output rather than by the experimenter's guess about grouping Quotation withheld because this record is not cleared for redistribution. a masking step is used to strip away everything except the contested store, so that what remains is the thing being measured Quotation withheld because this record is not cleared for redistribution. a post-stimulus retention cue splits encoding from maintenance, so the modality-specific and modality-general phases can be measured in the same trial Quotation withheld because this record is not cleared for redistribution. a warning against reading a box diagram as an ontology: the boxes record the order in which information arrives, not places where memory exclusively lives Quotation withheld because this record is not cleared for redistribution. the paper demonstrates its own subject matter on the reader inside a parenthesis, using sentence structure to load the reader's working memory Quotation withheld because this record is not cleared for redistribution. a mathematical model separates two stores by assuming their contributions are independent, S+L−SL; the paper then reports that the data fit a complementary-resource assumption better Quotation withheld because this record is not cleared for redistribution. a patient dissociation is used to identify which store a response came from: Korsakoff amnesia spares the last few serial positions in immediate recall and not in delayed recall Quotation withheld because this record is not cleared for redistribution. decay is reconceived from a gradual fade to a sudden collapse at a trial-varying point, which changes what an averaged curve is evidence for Quotation withheld because this record is not cleared for redistribution. rehearsal is split into two kinds so that the standard control is shown to be incomplete: articulatory suppression stops the covert-speech kind and leaves the attention-based kind untouched Quotation withheld because this record is not cleared for redistribution. Both of the properties on which the short-term/long-term distinction rests remain controversial what this limits: the short-term/long-term distinction itself: BOTH PROPERTIES OF SHORT-TERM MEMORY ARE STILL CONTROVERSIAL - the distinction rests on two properties, neither settled Quotation withheld because this record is not cleared for redistribution. Decay in particular is left unresolved by the review what this limits: the decay half of the distinction specifically: THE QUESTION OF DECAY IS STILL PRETTY MUCH OPEN TO DEBATE, WHEREAS THERE IS GROWING SUPPORT FOR A CHUNK CAPACITY LIMIT Quotation withheld because this record is not cleared for redistribution. How time limits and chunk capacity limits jointly govern recall is not settled what this limits: any account of recall that invokes both limits: THE EXACT WAY IN WHICH THESE LIMITS WORK TOGETHER IS NOT YET CLEAR Quotation withheld because this record is not cleared for redistribution. the decay claim is still not secure after fifty years of work what this limits: every decay-based explanation in the review: THE CONCEPT OF DECAY IS NOT YET ON VERY FIRM GROUND AND WARRANTS FURTHER STUDY Quotation withheld because this record is not cleared for redistribution. the two limits cannot be cleanly separated even when both are granted what this limits: the content of the term short-term memory: WE CAN SAY THAT SHORT-TERM MEMORY EXISTS, BUT OFTEN WITHOUT GREAT CLARITY AS TO WHETHER THE LIMIT IS A TIME LIMIT OR A CHUNK CAPACITY LIMIT - the category is used while its individuating property is undetermined Quotation withheld because this record is not cleared for redistribution. the number of chunks used in recall is not usually observable, so the capacity number is not directly measurable what this limits: every capacity number in the literature: BECAUSE IT IS NOT USUALLY CLEAR WHAT CHUNKS HAVE BEEN USED IN RECALL, IT IS NOT CLEAR HOW MANY CHUNKS CAN BE RETAINED AND WHETHER THE NUMBER IS TRULY FIXED Quotation withheld because this record is not cleared for redistribution. a design in which the grouping variable is also the grouping measure — Vogel et al. split participants by the very signal being compared what this limits: the Vogel et al. individual-differences result: THE SEPARATION OF PARTICIPANTS INTO HIGH VERSUS LOW SPAN WAS BASED ON THE CDA ALSO - the grouping variable is the measure under test Quotation withheld because this record is not cleared for redistribution. the interpretation of the working-memory/aptitude correlation, the field's main reason for using the term, is unsettled what this limits: the field's main justification for the term working memory: MORE RESEARCH IS NEEDED ON EXACTLY WHAT WE LEARN FROM THE HIGH CORRELATION BETWEEN WORKING MEMORY AND INTELLECTUAL APTITUDES Quotation withheld because this record is not cleared for redistribution. a proposed decay experiment is conceded in advance to be reinterpretable as interference, so the design cannot settle the question what this limits: the proposed decay experiment: SUCH RESULTS MIGHT BE ACCOUNTED FOR ALTERNATIVELY AS THE RESULT OF INTERFERENCE ... WITHOUT THE NEED TO INVOKE DECAY - conceded before it is run Quotation withheld because this record is not cleared for redistribution. how the capacity limit and the time limit combine is left unresolved what this limits: the same scope as limitations[2] - how the two limits combine Quotation withheld because this record is not cleared for redistribution. the practical demand that created the modern short-term memory task — operators holding rapid dot-dash series, then subscriber numbers — which the paper offers as the ecological source of the nonsense-syllable laboratory Quotation withheld because this record is not cleared for redistribution. the cell assembly, imported to give short-term memory a physical criterion independent of awareness Quotation withheld because this record is not cleared for redistribution. the independence rule for a union of two events, S+L−SL, used to split a single recall probability into contributions from two stores Quotation withheld because this record is not cleared for redistribution. a NON-STUDIED control condition, against list words that were each presented exactly four times; what varied was only how many of those presentations were paired eliminates: sheer exposure frequency as the explanation of recall - every studied word is seen the same number of times, so a difference between conditions must come from the PAIRING rather than from familiarity · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. long-term memory long-term memory Quotation withheld because this record is not cleared for redistribution. working memory working memory Quotation withheld because this record is not cleared for redistribution. focus of attention focus of attention Quotation withheld because this record is not cleared for redistribution. central executive central executive Quotation withheld because this record is not cleared for redistribution. articulatory suppression articulatory suppression Quotation withheld because this record is not cleared for redistribution. recency effect recency effect Quotation withheld because this record is not cleared for redistribution. contralateral delay activity contralateral delay activity Quotation withheld because this record is not cleared for redistribution. short-term → working memory NO BOUNDARY CAN BE DRAWN - the CONCEPTUAL form of the negative relation / This is a claim about the concepts themselves, not about this study's data. No edge type can express it: all eight assert that a relation HOLDS. From the paper that also says short-term memory exists 'without clarity as to whether the limit is time or capacity'. Here the field's two central terms are said to have no clear boundary AT ALL from one influential standpoint - and our corpus separately carries edges that treat them as distinct things. Quotation withheld because this record is not cleared for redistribution. working memory — three discrepant definitions named together in one sentence: short-term memory applied to cognitive tasks; a multi-component system; the use of attention to manage short-term memory Quotation withheld because this record is not cleared for redistribution. working memory — Cowan's own sense: short-term memory plus the mechanisms that make use of it Quotation withheld because this record is not cleared for redistribution. working memory — Engle's sense: the attention-related aspects only, which redefines short-term memory in the same move Quotation withheld because this record is not cleared for redistribution. working memory — Miller, Galanter and Pribram's original functional sense: memory as it is used to plan and carry out behaviour, with no distinction from short-term memory Quotation withheld because this record is not cleared for redistribution. short-term memory — broader than primary memory, because a temporarily accessible idea need not have been conscious Quotation withheld because this record is not cleared for redistribution. short-term memory — as a neural sense: an active cell assembly, whether or not the individual is aware of it Quotation withheld because this record is not cleared for redistribution. rehearsal — covert verbal (articulatory) versus attention-based; only the first is stopped by articulatory suppression Quotation withheld because this record is not cleared for redistribution. rehearsal — as even-paced covert articulation versus as a grouping operation; participants asked what they did overwhelmingly reported grouping Quotation withheld because this record is not cleared for redistribution. capacity limit — as duration (items decay from storage over time) versus as number (a ceiling on how many items are held); the paper treats these as the two separable ways stores could differ Quotation withheld because this record is not cleared for redistribution. decay — as gradual degradation of trace quality versus as sudden collapse at a trial-varying moment Quotation withheld because this record is not cleared for redistribution. attention in working memory — its control versus its scope; Cowan et al. treat these as independent contributors to aptitude Quotation withheld because this record is not cleared for redistribution. forgetting — as inevitable decay versus as interference during the retention interval, the split that founded the unitary-memory tradition Quotation withheld because this record is not cleared for redistribution. claim_type: critique a second reading DIFFERING from the recorded value. Recorded: existence — note: a second reading, verbatim-verified, disagreeing with the value on the record. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. method: conceptual review — assembles experimental evidence around a contamination framework, naming for each claimed limit the confounds that must be overcome before the evidence counts a second reading DIFFERING from the recorded value. Recorded: review — note: a second reading, verbatim-verified, disagreeing with the value on the record. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. phenomenon: memory over the short term — its duration limit, its capacity limit, and its relation to attention a second reading DIFFERING from the recorded value. Recorded: short-term, long-term and working memory distinctions — note: a second reading, verbatim-verified, disagreeing with the value on the record. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. builds-on → Cowan's own earlier theoretical framework (Cowan 1988, 1995, 1999, 2001, 2005) Quotation withheld because this record is not cleared for redistribution. extends → Baddeley and Hitch's multi-component working memory model Quotation withheld because this record is not cleared for redistribution. contradicts → Engle (2002) definition reserving 'working memory' for attention-related aspects of short-term memory differs in: the SCOPE of the term: some researchers WOULD LIKE TO RESERVE THE TERM WORKING MEMORY TO REFER ONLY TO THE ATTENTION-RELATED ASPECTS OF SHORT-TERM MEMORY, where this paper uses it more broadly Quotation withheld because this record is not cleared for redistribution. critiques → the field's use of the term working memory, which the paper diagnoses as a usage dispute rather than a substantive disagreement differs in: nothing substantive - the paper says so: THIS, HOWEVER, IS NOT SO MUCH A DEBATE ABOUT SUBSTANCE, BUT RATHER A SLIGHTLY CONFUSING DISCREPANCY IN THE USAGE OF TERMS Quotation withheld because this record is not cleared for redistribution. critiques → Miller's magical number seven, on the authority of Miller's own later account of it differs in: whether the original number was ever a claim: HIS AUTOBIOGRAPHICAL ESSAY INDICATES THAT HE WAS NEVER VERY SERIOUS ABOUT THE NUMBER SEVEN; IT WAS A RHETORICAL DEVICE THAT HE USED TO TIE TOGETHER THE OTHERWISE UNRELATED STRANDS OF HIS RESEARCH FOR A TALK Quotation withheld because this record is not cleared for redistribution. contradicts → the unitary-memory tradition (Melton, Crowder, Nairne, Lewandowsky and others) which holds that one set of principles covers memory at every timescale Quotation withheld because this record is not cleared for redistribution. extends → Baddeley and Hitch's multi-component working memory, kept as a shape but not as a component list Quotation withheld because this record is not cleared for redistribution. critiques → Cocchini et al.'s conclusion that dissimilar memory loads do not interfere, which would argue against a general cross-domain store differs in: THE AXIS IS INTERFERENCE BETWEEN DISSIMILAR LISTS. 'Quotation withheld.' The critique targets an INFERENCE rather than a measurement: the interference result is not disputed, the conclusion drawn from it is. Quotation withheld because this record is not cleared for redistribution. critiques → Vogel et al. (2005), for splitting participants on the same signal used as the dependent measure differs in: circularity in the split: THE SEPARATION OF PARTICIPANTS INTO HIGH VERSUS LOW SPAN WAS BASED ON THE CDA ALSO, AND THE TASK USED TO MEASURE THE CDA INEVITABLY REQUIRED SELECTIVE ATTENTION ON EVERY TRIAL - the grouping variable is the dependent measure Quotation withheld because this record is not cleared for redistribution. supports → Broadbent's and Cowan's revision of the capacity limit from seven to three or four units Quotation withheld because this record is not cleared for redistribution. episodic buffer —same-as→ focus of attention episodic buffer focus of attention strength of the relation: 0.6 — the record marks this weak Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the episodic buffer is 'POSSIBLY the same as' the focus of attention 'or at least a closely similar concept'; same-as is hedged. Quotation withheld because this record is not cleared for redistribution. phonological loop —part-of→ activated memory phonological loop activated memory strength of the relation: 0.85 Quotation withheld because this record is not cleared for redistribution. focus of attention —part-of→ activated memory focus of attention activated memory strength of the relation: 0.95 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span writes 'the activated information', not 'activated memory'. Quotation withheld because this record is not cleared for redistribution. short-term memory —part-of→ long-term memory short-term memory long-term memory strength of the relation: 0.85 Quotation withheld because this record is not cleared for redistribution. EVIDENCE CORRECTED (RE-GROUNDED) · temporal decay —distinct-from→ chunk capacity limits temporal decay chunk capacity limits EVIDENCE CORRECTED: RE-GROUNDED. The old span named neither. This sentence names both and states them as TWO fundamental ways of differing - which is exactly the distinctness the edge records. — strength of the relation: 0.95 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span lists 'temporal decay' and 'chunk capacity limits' as the two ways long- and short-term memory could differ; it does not say the two are distinct from each other. Quotation withheld because this record is not cleared for redistribution. temporal distinctiveness —contrasted-with→ temporal decay temporal distinctiveness temporal decay strength of the relation: 0.95 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span writes 'decay', not 'temporal decay'. Quotation withheld because this record is not cleared for redistribution. interference —contrasted-with→ temporal decay interference temporal decay differs in: THE AXIS IS WHAT HAPPENS DURING THE RETENTION INTERVAL, AND THE DECIDING OBSERVATION IS IN THE SAME SENTENCE. 'Quotation withheld.' Decay predicts a monotone loss; the improvement case is what decay cannot produce, so the two are separated by a possible observation rather than by definition. · strength of the relation: 0.95 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span writes 'decay of memory', not 'temporal decay'. Quotation withheld because this record is not cleared for redistribution. chunk capacity limits —distinct-from→ memory span chunk capacity limits memory span strength of the relation: 0.9 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span says a span of seven items does not guarantee seven separate chunks; distinct-from is loose. Quotation withheld because this record is not cleared for redistribution. covert verbal rehearsal —distinct-from→ attention-based rehearsal covert verbal rehearsal attention-based rehearsal strength of the relation: 0.95 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but endpoint by anaphora: 'that type of rehearsal' = attention-based reactivation (previous sentence). Quotation withheld because this record is not cleared for redistribution. control of attention —distinct-from→ scope of attention control of attention scope of attention strength of the relation: 0.9 Quotation withheld because this record is not cleared for redistribution. control of attention —operationalised-as→ antisaccade task control of attention antisaccade task status: ATTRIBUTED TO A NAMED STUDY, AND THIS IS A REVIEW, so reporting the field's operationalisations IS the paper's content: 'Quotation withheld correlates well with the ability to inhibit the natural tendency to look toward a suddenly appearing stimulus'. The edge is right about the field and silent about whose measurement it is - which is what this field records. · strength of the relation: 0.75 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the antisaccade task measures the ability to inhibit a look; 'control of attention' is not in the span, so operationalised-as is loose. Quotation withheld because this record is not cleared for redistribution. working memory —operationalised-as→ contralateral delay activity working memory contralateral delay activity status: ATTRIBUTED TO A NAMED STUDY IN A REVIEW: 'VOGEL ET AL. Quotation withheld' The operationalisation is Vogel et al.'s and the reporting is Cowan's. · strength of the relation: 0.85 Quotation withheld because this record is not cleared for redistribution. central executive —part-of→ working memory central executive working memory strength of the relation: 0.9 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but endpoint by anaphora: 'an influential model' = the working memory model (previous sentences); part-of is loose. Quotation withheld because this record is not cleared for redistribution. short-term memory —part-of→ working memory short-term memory working memory strength of the relation: 0.9 Quotation withheld because this record is not cleared for redistribution. chunk capacity limits —operationalised-as→ running memory span chunk capacity limits running memory span strength of the relation: 0.85 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but running memory span makes grouping IMPOSSIBLE, which is why it bears on chunk capacity; operationalised-as is loose. Quotation withheld because this record is not cleared for redistribution. sensory memory —distinct-from→ categorical memory sensory memory categorical memory strength of the relation: 0.9 Quotation withheld because this record is not cleared for redistribution. interaction effect interaction effect Quotation withheld because this record is not cleared for redistribution. mean mean Quotation withheld because this record is not cleared for redistribution. M M Quotation withheld because this record is not cleared for redistribution. significant interaction significant interaction Quotation withheld because this record is not cleared for redistribution. S+L−SL S+L−SL Quotation withheld because this record is not cleared for redistribution. correlate correlate Quotation withheld because this record is not cleared for redistribution. mathematical model mathematical model Quotation withheld because this record is not cleared for redistribution. Reviewing the evidence, the chapter concludes that the chunk capacity limit is well supported whereas temporal decay remains unsettled. Quotation withheld because this record is not cleared for redistribution. Later studies suggested the capacity limit is more typically three or four units. Quotation withheld because this record is not cleared for redistribution. Chen and Cowan (2005) showed the time limit and the chunk capacity limit in short-term memory are separate. WHOSE RESULT THIS IS: Chen and Cowan (2005) showed - THE EVIDENCE FOR THIS ROW ATTRIBUTES THE WORK TO SOMEBODY ELSE. The cited name is the subject of a result verb and the sentence carries no first-person marker, so the finding, quantity or definition is the CITED author's rather than this record's. The claim is true and the quote is verbatim - what is wrong is whose result the corpus says it is. Nothing edited, nothing withdrawn: the row is marked so a page can name the source. Where the record is a commentary or review, its claim relations layer usually already names the paper being discussed - the record knows at one layer and forgets at another. Quotation withheld because this record is not cleared for redistribution. The results suggest there could be decay in short-term memory. Quotation withheld because this record is not cleared for redistribution. The focus of attention again held three to four items for both unimodal visual and bimodal stimuli. Quotation withheld because this record is not cleared for redistribution. Retention after the cue showed dual-task effects that were not modality-specific. Quotation withheld because this record is not cleared for redistribution. Because the four-item limit appears in vision and audition despite very different sensory persistence, it is a capacity limit and not a transfer-rate limit. Quotation withheld because this record is not cleared for redistribution. Individual capacity varies from roughly two to six items in adults and predicts cognitive aptitude. Quotation withheld because this record is not cleared for redistribution. Whether short-term and working memory differ may be a matter of semantics. Quotation withheld because this record is not cleared for redistribution. the capacity limit is more typically three or four units, not seven Quotation withheld because this record is not cleared for redistribution. Miller was never serious about seven — by his own later account it was a rhetorical device to tie unrelated strands of research together for a talk Quotation withheld because this record is not cleared for redistribution. individual capacity ranges from about two to six items in adults and is a strong correlate of cognitive aptitude Quotation withheld because this record is not cleared for redistribution. visual and auditory experiments give a common four-item limit despite very different transfer times, so it is best viewed as a capacity rather than a rate limit Quotation withheld because this record is not cleared for redistribution. the focus of attention held between three and four items for unimodal visual or bimodal stimuli Quotation withheld because this record is not cleared for redistribution. after initial encoding, working memory storage over several seconds may occur abstractly, in the focus of attention Quotation withheld because this record is not cleared for redistribution. Lewandowsky et al. found no decay across response intervals from 400 to 1600 ms under articulatory suppression Quotation withheld because this record is not cleared for redistribution. when recall timing had to match presentation timing, a long first half of the response intervals hurt recall (.67 vs .70) Quotation withheld because this record is not cleared for redistribution. the Peterson delay effect is absent on the first few trials and when delays are tested in separate blocks WHOSE RESULT THIS IS: ested in separate trial blocks ( Turvey et al., 1970 ; Greene, 1996 ). - THE RESULT IS THE CITED WORK'S AND THE CITATION TRAILS THE SENTENCE. A plain noun phrase is the subject of the result verb and the attribution is a bracket at the end - 'Neurons within the dorsal temporal pole demonstrated match suppression ... ( Ng et al., 2014 )'. Neither of my two cues could see this shape, so THESE WERE READ RATHER THAN SWEPT: 36 candidates of the shape, 20 read, 11 marked, 9 refused. THE REFUSALS ARE WHY IT IS NOT A CUE: 'Aru2012 findings[16]' is the paper's OWN result with a citation supporting the localisation, 'Natan2015 findings[3]' ends 'consistent with previous findings ( Szymanski et al., 2009 )' where the main clause is the paper's, and 'Drieu2019' x3 are 'One cannot entirely rule out...' - a limitation, not an attribution. About 55% precise, which is a list to read and not a sweep to run. Quotation withheld because this record is not cleared for redistribution. an effect of test delay appeared within the first 5 s but not at longer delays Quotation withheld because this record is not cleared for redistribution. in delayed recall the amnesic patients show a deficit at all serial positions Quotation withheld because this record is not cleared for redistribution. when recalling from a long-term category such as US states, people produce items in spurts of about three Quotation withheld because this record is not cleared for redistribution. capacity estimated at about 3.5 chunks in every learning condition, with prior exposure changing the chunk composition rather than the number Quotation withheld because this record is not cleared for redistribution. in adults asked how they did a digit span task, grouping was by far the most common answer and saying the items to themselves was rarely mentioned Quotation withheld because this record is not cleared for redistribution. at four items per second in running span, instructions to rehearse make performance worse rather than better Quotation withheld because this record is not cleared for redistribution. the time limit and the chunk capacity limit are separable: long free-recall lists are governed by capacity, short strictly-scored serial lists by time Quotation withheld because this record is not cleared for redistribution. a random seven-digit list interfered with visual array comparison but the participant's own telephone number did not, and only when recited aloud Quotation withheld because this record is not cleared for redistribution. in schizophrenia the working memory deficit was in capacity, not in the control of attention — against the authors' own expectation Quotation withheld because this record is not cleared for redistribution. of the central executive functions, updating correlated with aptitudes while inhibition and shifting did not WHOSE RESULT THIS IS: Friedman et al. (2006) found - THE EVIDENCE FOR THIS ROW ATTRIBUTES THE WORK TO SOMEBODY ELSE. The cited name is the subject of a result verb and the sentence carries no first-person marker, so the finding, quantity or definition is the CITED author's rather than this record's. The claim is true and the quote is verbatim - what is wrong is whose result the corpus says it is. Nothing edited, nothing withdrawn: the row is marked so a page can name the source. Where the record is a commentary or review, its claim relations layer usually already names the paper being discussed - the record knows at one layer and forgets at another. Quotation withheld because this record is not cleared for redistribution. high-span individuals notice their own name in an ignored dichotic channel much less often than low-span individuals WHOSE RESULT THIS IS: Conway et al. (2001) found - THE EVIDENCE FOR THIS ROW ATTRIBUTES THE WORK TO SOMEBODY ELSE. The cited name is the subject of a result verb and the sentence carries no first-person marker, so the finding, quantity or definition is the CITED author's rather than this record's. The claim is true and the quote is verbatim - what is wrong is whose result the corpus says it is. Nothing edited, nothing withdrawn: the row is marked so a page can name the source. Where the record is a commentary or review, its claim relations layer usually already names the paper being discussed - the record knows at one layer and forgets at another. Quotation withheld because this record is not cleared for redistribution. in children too young for covert verbal rehearsal, even a plain digit span task correlates excellently with aptitudes Quotation withheld because this record is not cleared for redistribution. recognition of earlier list portions, but not the last few items, activated hippocampal areas associated with long-term retrieval WHOSE RESULT THIS IS: Talmi et al. (2005) found - THE EVIDENCE FOR THIS ROW ATTRIBUTES THE WORK TO SOMEBODY ELSE. The cited name is the subject of a result verb and the sentence carries no first-person marker, so the finding, quantity or definition is the CITED author's rather than this record's. The claim is true and the quote is verbatim - what is wrong is whose result the corpus says it is. Nothing edited, nothing withdrawn: the row is marked so a page can name the source. Where the record is a commentary or review, its claim relations layer usually already names the paper being discussed - the record knows at one layer and forgets at another. Quotation withheld because this record is not cleared for redistribution. a possible sudden collapse rather than gradual fade, seen for tone memory between 5 and 10 s with temporal distinctiveness controlled WHOSE RESULT THIS IS: Cowan et al. (1997a) found - THE EVIDENCE FOR THIS ROW ATTRIBUTES THE WORK TO SOMEBODY ELSE. The cited name is the subject of a result verb and the sentence carries no first-person marker, so the finding, quantity or definition is the CITED author's rather than this record's. The claim is true and the quote is verbatim - what is wrong is whose result the corpus says it is. Nothing edited, nothing withdrawn: the row is marked so a page can name the source. Where the record is a commentary or review, its claim relations layer usually already names the paper being discussed - the record knows at one layer and forgets at another. Quotation withheld because this record is not cleared for redistribution. free recall of word lists at every level of structure yielded an average of four to six chunks Quotation withheld because this record is not cleared for redistribution. long-term memory long-term memory Quotation withheld because this record is not cleared for redistribution. short-term memory short-term memory Quotation withheld because this record is not cleared for redistribution. working memory working memory Quotation withheld because this record is not cleared for redistribution. duration difference duration difference Quotation withheld because this record is not cleared for redistribution. capacity difference capacity difference Quotation withheld because this record is not cleared for redistribution. focus of attention focus of attention Quotation withheld because this record is not cleared for redistribution. chunk chunk Quotation withheld because this record is not cleared for redistribution. covert verbal rehearsal covert verbal rehearsal Quotation withheld because this record is not cleared for redistribution. articulatory suppression articulatory suppression Quotation withheld because this record is not cleared for redistribution. contralateral delay activity (CDA) contralateral delay activity (CDA) Quotation withheld because this record is not cleared for redistribution. Long-term memory Long-term memory Quotation withheld because this record is not cleared for redistribution. Short-term memory Short-term memory Quotation withheld because this record is not cleared for redistribution. primary memory primary memory Quotation withheld because this record is not cleared for redistribution. working memory working memory Quotation withheld because this record is not cleared for redistribution. A chunk A chunk Quotation withheld because this record is not cleared for redistribution. covert verbal rehearsal covert verbal rehearsal Quotation withheld because this record is not cleared for redistribution. articulatory suppression articulatory suppression Quotation withheld because this record is not cleared for redistribution. A duration difference A duration difference Quotation withheld because this record is not cleared for redistribution. A capacity difference A capacity difference Quotation withheld because this record is not cleared for redistribution. continual distractor list recall continual distractor list recall Quotation withheld because this record is not cleared for redistribution. running memory span running memory span Quotation withheld because this record is not cleared for redistribution. contralateral delay activity contralateral delay activity Quotation withheld because this record is not cleared for redistribution. temporal distinctiveness temporal distinctiveness Quotation withheld because this record is not cleared for redistribution. central executive central executive Quotation withheld because this record is not cleared for redistribution. scope scope Quotation withheld because this record is not cleared for redistribution. three or four units — the capacity limit of short-term storage once chunking is controlled Quotation withheld because this record is not cleared for redistribution. about 3.5 chunks — model-estimated capacity limit, constant across learning conditions, in the paired-word list study Quotation withheld because this record is not cleared for redistribution. approximately seven items — adult memory span, which the chapter argues overstates capacity because items are not separate chunks Quotation withheld because this record is not cleared for redistribution. four to six chunks — average number of chunks recalled across all structure conditions in the free-recall chunk study Quotation withheld because this record is not cleared for redistribution. M = .70 vs M = .67 — recall proportion with short versus long first-half response intervals when recall timing had to match presentation, the pattern taken as evidence of decay Quotation withheld because this record is not cleared for redistribution. M =.71 vs M =.74 — recall proportion with short versus long first-half intervals under free-paced recall, where no decay cost appeared Quotation withheld because this record is not cleared for redistribution. 5 s — delay window within which a proactive-interference-free study found a genuine test-delay effect Quotation withheld because this record is not cleared for redistribution. two to six items — range of individual adult capacity Quotation withheld because this record is not cleared for redistribution. three or four units — the revised chunk capacity limit Quotation withheld because this record is not cleared for redistribution. about seven items — adult memory span, which is not the same as the chunk capacity Quotation withheld because this record is not cleared for redistribution. about two to six items — range of individual capacity in adults Quotation withheld because this record is not cleared for redistribution. three items — Broadbent's error-free limit, and the size of recall spurts from a long-term category Quotation withheld because this record is not cleared for redistribution. about 3.5 chunks — capacity estimated by the chunk model in every learning condition Quotation withheld because this record is not cleared for redistribution. four to six chunks — average recalled across all levels of list structure in Tulving and Patkau Quotation withheld because this record is not cleared for redistribution. 400 and 1600 ms — range of inter-item response time over which Lewandowsky et al. found no decay Quotation withheld because this record is not cleared for redistribution. M = .70 versus M = .67 — recall with a short versus long first half when recall timing had to match presentation timing Quotation withheld because this record is not cleared for redistribution. M =.71 versus M =.74 — the same comparison when participants recalled at their own pace — the effect reverses Quotation withheld because this record is not cleared for redistribution. up to 18 s — the filled delay in the Peterson and Peterson trigram procedure Quotation withheld because this record is not cleared for redistribution. first 5 s — window in which a delay effect survived the one-trial-per-participant control Quotation withheld because this record is not cleared for redistribution. between 5 and 10 s — delay range over which tone memory appeared to collapse suddenly Quotation withheld because this record is not cleared for redistribution. about 4s — auditory sensory memory decay period in Darwin et al., against about 1 s for Sperling's visual case Quotation withheld because this record is not cleared for redistribution. 4 per second — presentation rate in running span fast enough to make rehearsal impossible Quotation withheld because this record is not cleared for redistribution. 0.5 s and 2 s — the short and long inter-digit intervals in the Cowan and Aubuchon timing design Quotation withheld because this record is not cleared for redistribution. 31 days — longest retention interval in Ebbinghaus's self-experiments Quotation withheld because this record is not cleared for redistribution. more than 200 subscribers — the Lowell telephone exchange that introduced numbers, in 1879 Quotation withheld because this record is not cleared for redistribution. 0, 1, 2, or 4 prior word–word pairings — the learning levels used to vary chunk composition Quotation withheld because this record is not cleared for redistribution. seven printed digits — list length in the Cowan and Aubuchon timing experiment Quotation withheld because this record is not cleared for redistribution. 75% — proportion of trials on which the probed set was the cued one in Gold et al. Quotation withheld because this record is not cleared for redistribution. memory span memory span Quotation withheld because this record is not cleared for redistribution. inter-digit blank intervals inter-digit blank intervals Quotation withheld because this record is not cleared for redistribution. response cue response cue Quotation withheld because this record is not cleared for redistribution. presentation rate presentation rate Quotation withheld because this record is not cleared for redistribution. memory load memory load Quotation withheld because this record is not cleared for redistribution. contralateral delay activity contralateral delay activity Quotation withheld because this record is not cleared for redistribution. serial position serial position Quotation withheld because this record is not cleared for redistribution. cognitive aptitude cognitive aptitude Quotation withheld because this record is not cleared for redistribution.

The full text of this article is not redistributable, so it is not included here. The evidence sentences are withheld with it; the tag layer and its grounding status remain. The article itself is at https://doi.org/10.1016/s0079-6123(07)00020-9.