Working Memory-Related Hippocampal Deactivation Interferes with Long-Term Memory Formation

Working Memory-Related Hippocampal Deactivation Interferes with Long-Term Memory Formation Axmacher Elger Fell

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.

Working Memory-Related Hippocampal Deactivation Interferes with Long-Term Memory Formation 2009 Journal of Neuroscience 10.1523/jneurosci.5277-08.2009 Axmacher2009-working-memory-related-hippocampal-deactivat Axmacher2009-working-memory-related-hippocampal-deactivat.txt, the project's staged copy dc1e227487b54f25fb85d2838232c5f6f4f57a4e not-redistributable neuroscience

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 whether working-memory-related hippocampal activity patterns during a word list predict long-term encoding, and explain the primacy effect Quotation withheld because this record is not cleared for redistribution. which form of working memory supports the items at the start of a list Quotation withheld because this record is not cleared for redistribution. shows hippocampal deactivation at the first list position predicts that the item will be forgotten, linking a working-memory signature to long-term encoding failure Quotation withheld because this record is not cleared for redistribution. mechanism Quotation withheld because this record is not cleared for redistribution. working memory-long-term memory interaction in the hippocampus Quotation withheld because this record is not cleared for redistribution. fmri fmri Quotation withheld because this record is not cleared for redistribution. mechanisms Early list items processed by single-item, hippocampus-independent working memory leave the hippocampus deactivated, so no encoding into long-term memory takes place and the item is forgotten Quotation withheld because this record is not cleared for redistribution. mechanisms When multi-item, hippocampus-dependent working memory is recruited instead at early list positions, the same hippocampal engagement supports long-term encoding Quotation withheld because this record is not cleared for redistribution. mechanisms Items early in the list are rehearsed more, and later items interfere with earlier ones, which is why the primacy advantage arises and why hippocampal engagement stops predicting memory later in the list status: the paper's own words: we suggest that Quotation withheld because this record is not cleared for redistribution. working memory is divided by neural signature: hippocampal activation for multiple items, deactivation for individual items Quotation withheld because this record is not cleared for redistribution. positive DC slopes in the iEEG are taken to correspond to BOLD deactivations, which is what lets the two datasets be read together Quotation withheld because this record is not cleared for redistribution. items at the start of a list are proposed to be rehearsed more, linking the primacy effect to working memory Quotation withheld because this record is not cleared for redistribution. successful early items are encoded in runs: encoding the first raises the conditional probability of encoding the second and third Quotation withheld because this record is not cleared for redistribution. only the hemisphere contralateral to the seizure focus was analysed, in patients with confirmed unilateral sclerosis Quotation withheld because this record is not cleared for redistribution. No behavioural measure of working-memory performance was collected, so it is unknown which items were actually held in WM. what this limits: every claim about what was held in working memory: THERE IS NO BEHAVIORAL MONITORING OF WM PERFORMANCE IN THE CURRENT EXPERIMENT Quotation withheld because this record is not cleared for redistribution. The link between the observed activity and actual long-term memory consequences of WM performance cannot be established. what this limits: any claim about what was held in mind: WE DO NOT KNOW FOR SURE HOW LTM IS ACTUALLY AFFECTED BY THE WM PERFORMANCE FOR SPECIFIC ITEMS, I.E., IT IS UNKNOWN WHICH ITEMS ARE ACTUALLY MAINTAINED IN WM Quotation withheld because this record is not cleared for redistribution. The proposed link to persistent activity recorded in vitro is acknowledged as speculative. what this limits: the link to the in vitro literature: THIS ANALOGY BETWEEN ELECTROPHYSIOLOGICAL DATA IN VITRO AND THE NEURAL PATTERNS DURING WM MAINTENANCE MUST REMAIN SPECULATIVE AT THE CURRENT STAGE Quotation withheld because this record is not cleared for redistribution. Position-specific subsequent-memory analyses were statistically underpowered. what this limits: the position-specific null: THIS LACK OF AN EFFECT IS RELATED TO THE REDUCED STATISTICAL POWER FOR POSITION-SPECIFIC ANALYSES OF SUBSEQUENT MEMORY EFFECTS Quotation withheld because this record is not cleared for redistribution. the two populations differ in recall performance despite a shorter recall phase for the fMRI group what this limits: any comparison between the two populations: Quotation withheld Quotation withheld because this record is not cleared for redistribution. the paradigms differ between the two versions in presentation time, interstimulus interval, block count and recall duration what this limits: any comparison between the fMRI and intracranial versions: they differ in presentation time (1000 ms, since goggles REQUIRED LONGER PRESENTATION TIMES), interstimulus interval and block count Quotation withheld because this record is not cleared for redistribution. patients showed both primacy and recency; the fMRI subjects showed no recency at all what this limits: any serial-position claim carried across the two groups: NO RECENCY EFFECTS AT THE END of the list were found in the fMRI subjects, against primacy AND recency in the patients Quotation withheld because this record is not cleared for redistribution. the whole-brain effects did not survive correction for multiple comparisons and are reported at the cluster level instead what this limits: the whole-brain effects: THESE EFFECTS DID NOT SURVIVE CORRECTION FOR MULTIPLE COMPARISONS, BUT DID HAVE A LARGE EXTENSION OF 60 VOXELS FOR THE SERIAL POSITION CONTRAST OF SUBSEQUENTLY FORGOTTEN WORDS Quotation withheld because this record is not cleared for redistribution. several supporting results are trends at p between 0.05 and 0.07, labelled as such what this limits: the subsequently-forgotten results, with their exact values: WE FOUND A TREND FOR AN EFFECT OF POSITION (F (2,36) = 2.943; p = 0.070) AND FOR A LINEAR EFFECT OF POSITION (F (1,18) = 4.249; p = 0.054) Quotation withheld because this record is not cleared for redistribution. a position-specific null is attributed to reduced statistical power duplicate of: limitations[3] · what this limits: the same scope as limitations[3] - the position-specific analyses, underpowered Quotation withheld because this record is not cleared for redistribution. effects appear in the left hippocampus and not the right, without an account of why what this limits: any bilateral hippocampal claim: FOR THE RIGHT HIPPOCAMPUS, A THREE-WAY ANOVA DID NOT REVEAL AN EFFECT OF, OR INTERACTION WITH, THE FACTOR POSITION (ALL p > 0.3), BUT A SIGNIFICANT MAIN EFFECT OF MEMORY (F (1,18) = 8.898; p < 0.01) Quotation withheld because this record is not cleared for redistribution. the required rote strategy is verified only by asking participants afterwards what this limits: the rote-strategy assumption: ALL SUBJECTS WERE REQUIRED TO USE A ROTE STRATEGY ... AVOIDING MEMORY AIDS - instructed, and verified only by asking afterwards Quotation withheld because this record is not cleared for redistribution. serial position curves and the primacy effect, imported as the behavioural handle on a neural question Quotation withheld because this record is not cleared for redistribution. a group of healthy control subjects investigated with fMRI alongside the patient recordings eliminates: the hippocampal working-memory effect being a property of the epileptic brains the depth electrodes are implanted in · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. working memory is subdivided here by NEURAL SIGNATURE rather than by task or capacity: hippocampus-dependent when multiple items are held, hippocampus-independent when one is Quotation withheld because this record is not cleared for redistribution. a deactivation is used as a positive finding - the absence of hippocampal engagement is the marker, not a null Quotation withheld because this record is not cleared for redistribution. effects of position and LINEAR effects of position are distinguished explicitly, with the a priori assumption of the second stated: that if there are effects across positions, these should be linear Quotation withheld because this record is not cleared for redistribution. the DC potential and the BOLD signal are treated as two measures of one underlying quantity, on the strength of a cited correspondence rather than a within-subject comparison Quotation withheld because this record is not cleared for redistribution. claim_type: mechanism a second reading reaching the same value. Recorded: mechanism — note: a second reading reaching the same value from a different sentence. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. method: intracranial EEG DC potentials in 8 epilepsy patients reanalysed from a 1999 study, plus new fMRI in 19 healthy subjects on the same word-list paradigm a second reading DIFFERING from the recorded value. Recorded: fmri — 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: the primacy effect and its hippocampal correlates a second reading DIFFERING from the recorded value. Recorded: working memory-long-term memory interaction in the hippocampus — 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 → Fernández et al., 1999 Quotation withheld because this record is not cleared for redistribution. extends → Axmacher et al., 2007 Quotation withheld because this record is not cleared for redistribution. builds-on → Atkinson and Shiffrin (1968) Quotation withheld because this record is not cleared for redistribution. extends → the authors' own Sternberg study distinguishing hippocampus-dependent from hippocampus-independent working memory Quotation withheld because this record is not cleared for redistribution. extends → Fernandez et al. 1999, whose iEEG data are reanalysed here for a different question Quotation withheld because this record is not cleared for redistribution. supports → the rehearsal account of the primacy effect Quotation withheld because this record is not cleared for redistribution. supports → fMRI findings that working-memory maintenance facilitates long-term encoding Quotation withheld because this record is not cleared for redistribution. deactivation —causes→ long-term memory deactivation long-term memory strength of the relation: 0.8 · note: predicts FAILURE of encoding; the sign matters and the schema records only the pair Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span says hippocampal deactivation predicts failure of encoding; the edge carries no direction. Quotation withheld because this record is not cleared for redistribution. working memory —causes→ long-term memory working memory long-term memory status: ATTRIBUTED TO NAMED WORK AND OFFERED AS AN IDEA: 'IT HAS BEEN PROPOSED THAT the primacy effect is due to the fact that items at early list positions undergo more extensive rehearsal than items presented later (Rundus, 1971; Brodie and Murdock, 1977). THIS IDEA SUGGESTS a link between LTM formation and rehearsal processes'. The relation is real and this paper builds on it; it belongs to two papers from the 1970s. · strength of the relation: 0.8 · note: via rehearsal, which is the proposed account of the primacy effect Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span says the idea 'SUGGESTS A LINK' between LTM formation and rehearsal attributed to working memory; causes is hedged. Quotation withheld because this record is not cleared for redistribution. primacy —contrasted-with→ recency primacy recency strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span reports only the recency effect; the contrast with primacy is not in it. Quotation withheld because this record is not cleared for redistribution. ANOVA ANOVA Quotation withheld because this record is not cleared for redistribution. ANOVA ANOVA Quotation withheld because this record is not cleared for redistribution. sphericity correction sphericity correction Quotation withheld because this record is not cleared for redistribution. t-test t-test Quotation withheld because this record is not cleared for redistribution. t-test t-test Quotation withheld because this record is not cleared for redistribution. t-test t-test Quotation withheld because this record is not cleared for redistribution. regression regression Quotation withheld because this record is not cleared for redistribution. general linear model general linear model Quotation withheld because this record is not cleared for redistribution. multiple-comparison correction multiple-comparison correction Quotation withheld because this record is not cleared for redistribution. statistical-significance statistical-significance Quotation withheld because this record is not cleared for redistribution. repeated measures ANOVA with Huynh-Feldt correction repeated measures ANOVA with Huynh-Feldt correction Quotation withheld because this record is not cleared for redistribution. linear regression of slopes across positions linear regression of slopes across positions Quotation withheld because this record is not cleared for redistribution. conditional versus unconditional probability conditional versus unconditional probability Quotation withheld because this record is not cleared for redistribution. cluster-level significance cluster-level significance Quotation withheld because this record is not cleared for redistribution. As expected, more items were successfully encoded at the beginning of the list than at later positions. Quotation withheld because this record is not cleared for redistribution. Early-list words that were later forgotten were accompanied by hippocampal deactivation signatures in both fMRI and intracranial EEG. Quotation withheld because this record is not cleared for redistribution. No such deactivation appeared for later list items or for items that were successfully remembered. Quotation withheld because this record is not cleared for redistribution. In the intracranial data the linear decline of DC slope across list positions was significant only for subsequently forgotten items. Quotation withheld because this record is not cleared for redistribution. A second, independent general linear model reproduced the effect: left-hippocampal negative BOLD for forgotten items at position one was absent at position seven. Quotation withheld because this record is not cleared for redistribution. The position effect was left-lateralised: the right hippocampus showed a main effect of subsequent memory but no position effect. Quotation withheld because this record is not cleared for redistribution. For both later-remembered and later-forgotten items, activity was most pronounced in the anterior cingulate cortex. Quotation withheld because this record is not cleared for redistribution. Successful encoding at the start of a list came in runs: encoding the first word raised the probability of encoding the second. Quotation withheld because this record is not cleared for redistribution. The authors conclude that failing to engage hippocampus-dependent working memory early in a list causes the later encoding failure. Quotation withheld because this record is not cleared for redistribution. hippocampal deactivation at the first list position occurred for items that were later forgotten Quotation withheld because this record is not cleared for redistribution. no such deactivation appeared for items that were later remembered Quotation withheld because this record is not cleared for redistribution. DC slopes were positive early in the list for subsequently forgotten items and flattened later Quotation withheld because this record is not cleared for redistribution. only the first position differed significantly from zero Quotation withheld because this record is not cleared for redistribution. in the iEEG subjects, encoding the first item raised the conditional probability of encoding the second (0.35 to 0.51) Quotation withheld because this record is not cleared for redistribution. the patients showed a recency effect at the end of the list Quotation withheld because this record is not cleared for redistribution. the healthy fMRI group showed primacy but no recency at all Quotation withheld because this record is not cleared for redistribution. a second general linear model comparing the first and seventh positions reproduced the result Quotation withheld because this record is not cleared for redistribution. primacy effect primacy effect Quotation withheld because this record is not cleared for redistribution. linear effects of position linear effects of position Quotation withheld because this record is not cleared for redistribution. effects of position effects of position Quotation withheld because this record is not cleared for redistribution. hippocampus-dependent WM hippocampus-dependent WM Quotation withheld because this record is not cleared for redistribution. hippocampus-independent WM hippocampus-independent WM Quotation withheld because this record is not cleared for redistribution. list position list position Quotation withheld because this record is not cleared for redistribution. group group Quotation withheld because this record is not cleared for redistribution. DC slopes DC slopes Quotation withheld because this record is not cleared for redistribution. negative BOLD response negative BOLD response Quotation withheld because this record is not cleared for redistribution. hemisphere hemisphere Quotation withheld because this record is not cleared for redistribution. serial position serial position Quotation withheld because this record is not cleared for redistribution. subsequent memory subsequent memory Quotation withheld because this record is not cleared for redistribution. slope of the DC potential slope of the DC potential Quotation withheld because this record is not cleared for redistribution. Eight — epilepsy patients contributing intracranial EEG data Quotation withheld because this record is not cleared for redistribution. Nineteen — healthy participants in the fMRI experiment Quotation withheld because this record is not cleared for redistribution. 4.9% — proportion of EEG trials rejected as artefacts before analysis Quotation withheld because this record is not cleared for redistribution. 0.35 / 0.51 — unconditional versus conditional probability of encoding the second list item in the iEEG patients Quotation withheld because this record is not cleared for redistribution. smaller than 0.19 — maximum collinearity between the fMRI model regressors for adjacent list items (0 = orthogonal, 1 = collinear) Quotation withheld because this record is not cleared for redistribution. F (11,198) = 21.156; p < 10 −16 — effect of serial position on recall performance in the fMRI participants Quotation withheld because this record is not cleared for redistribution. 8 epilepsy patients and 19 healthy subjects — the two samples Quotation withheld because this record is not cleared for redistribution. 0.35 rising to 0.51 for the second item given the first — conditional encoding probability in patients Quotation withheld because this record is not cleared for redistribution. 4.9% of trials excluded for artefacts — iEEG data rejection Quotation withheld because this record is not cleared for redistribution. MNI 30/-15/-18 and -30/-15/-18 — the anatomically defined hippocampal regions of interest Quotation withheld because this record is not cleared for redistribution. recall phase 90 s for iEEG versus 60 s for fMRI, presentation 400 ms versus 1000 ms — the paradigm differences between the two versions value source note: THE VALUE DRAWS ON THE PAPER BEYOND THIS QUOTE: part of it is printed elsewhere in the paper, not in this quote ('presented in uppercase letters for a duration of 400 ms.'). The 90 s is the 60 s of this quote plus the 30 s it says the other version lasted longer. Quotation withheld because this record is not cleared for redistribution. p = 0.054, p = 0.070, p = 0.054 — the trends reported as consistent with the iEEG result 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/10.1523/jneurosci.5277-08.2009.