The Hippocampus Supports High-Resolution Binding in the Service of Perception, Working Memory and Long-Term Memory

The Hippocampus Supports High-Resolution Binding in the Service of Perception, Working Memory and Long-Term Memory Yonelinas

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.

The Hippocampus Supports High-Resolution Binding in the Service of Perception, Working Memory and Long-Term Memory 2013 Behavioural brain research 10.1016/j.bbr.2013.05.030 Yonelinas2013-the-hippocampus-supports-high-resolution-bin Yonelinas2013-the-hippocampus-supports-high-resolution-bin.txt, the project's staged copy 5724a51fed3571f290789687d3938f4678389162 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 why hippocampal patients are impaired on some working memory and perception tasks and not on others that look the same Quotation withheld because this record is not cleared for redistribution. whether the hippocampus supports complex high-resolution associative information across memory, working memory and perception Quotation withheld because this record is not cleared for redistribution. the discriminating variable is resolution, not association: hippocampal damage impairs tasks needing complex high-resolution bindings and spares those needing only low-resolution ones Quotation withheld because this record is not cleared for redistribution. mechanism Quotation withheld because this record is not cleared for redistribution. hippocampal high-resolution binding Quotation withheld because this record is not cleared for redistribution. review review Quotation withheld because this record is not cleared for redistribution. mechanisms Over long delays only the hippocampus can hold complex high-resolution bindings, so its necessity grows as the delay lengthens and other cortical stores drop out Quotation withheld because this record is not cleared for redistribution. mechanisms Cortical perceptual binding regions lose their representations as delay grows, presumably through interference from incoming sensory input, shifting the load onto the hippocampus Quotation withheld because this record is not cleared for redistribution. mechanisms The participant's strategy determines hippocampal recruitment: a single-feature search bypasses high-resolution hippocampal representations, whereas configural matching requires them Quotation withheld because this record is not cleared for redistribution. mechanisms Pattern separation in dentate gyrus and CA3 rapidly builds distinct representations of complex configurations, the substrate the account requires duplicate of: mechanisms[4] Quotation withheld because this record is not cleared for redistribution. pattern separation in dentate gyrus and CA3 allows rapid formation of distinct representations for complex configurations, which is the neurobiological form of the proposal Quotation withheld because this record is not cleared for redistribution. the same representations should be available without any long delay, which is why they could serve working memory and perception too Quotation withheld because this record is not cleared for redistribution. over short delays other cortical mechanisms can hold such bindings, so the hippocampal contribution only becomes visible as the delay grows Quotation withheld because this record is not cleared for redistribution. the domain matters as well as the resolution: simple associations need the hippocampus in long-term memory but not in working memory, and in perception only complex materials do Quotation withheld because this record is not cleared for redistribution. The account does not address semantic or remote long-term memory. what this limits: the scope of the account: HOW THE CURRENT IDEAS RELATE TO SEMANTIC LONG-TERM MEMORY OR REMOTE MEMORY is not addressed Quotation withheld because this record is not cleared for redistribution. The supporting computational simulations are preliminary and not yet shown to cover the working memory and perceptual literature. what this limits: the computational support: FUTURE WORK WILL BE NECESSARY TO DETERMINE IF THE SAME MODEL PARAMETERS ARE CAPABLE OF ACCURATELY ACCOUNTING FOR OTHER ASPECTS OF THE WORKING MEMORY AND PERCEPTUAL LITERATURE Quotation withheld because this record is not cleared for redistribution. Whether recollection in working memory can occur without the hippocampus is only indirectly supported. what this limits: the hippocampal-dependence claim for working-memory recollection: THE EXISTING RESULTS ARE NOT YET ENTIRELY CLEAR, BUT THERE IS INDIRECT EVIDENCE THAT RECOLLECTION IN WORKING MEMORY CAN ARISE INDEPENDENTLY OF THE HIPPOCAMPUS Quotation withheld because this record is not cleared for redistribution. whether high-resolution semantic or remote memories depend on the hippocampus is stated to be unclear, and they differ in being learned gradually what this limits: the same scope as limitations[0], with the difficulty named: INDIVIDUALS CAN ALSO DEVELOP HIGH-RESOLUTION SEMANTIC OR REMOTE MEMORIES, which the account does not cover Quotation withheld because this record is not cleared for redistribution. there is indirect evidence that recollection in working memory can arise without the hippocampus, which the account does not yet explain what this limits: the same scope as limitations[2] - the evidence against hippocampal necessity here is indirect Quotation withheld because this record is not cleared for redistribution. whether expertise turns a high-resolution feature into a low-resolution one is raised as an open question what this limits: the high/low resolution distinction itself: WHETHER EXPERTISE CAN INFLUENCE WHETHER A STIMULUS CHARACTERISTIC IS PROCESSED AS A HIGH OR LOW-RESOLUTION FEATURE is open Quotation withheld because this record is not cleared for redistribution. one supporting study is cited with the note that it provided no statistical comparisons what this limits: the small-set-size evidence specifically: Quotation withheld (EXPERIMENT 2, ALTHOUGH NOTE THAT NO STATISTICAL COMPARISONS WERE PROVIDED) Quotation withheld because this record is not cleared for redistribution. receiver operating characteristics used to split a perceptual judgement into two processes rather than score it as one Quotation withheld because this record is not cleared for redistribution. the continuous colour wheel, which separates precision from capacity where an accuracy score cannot Quotation withheld because this record is not cleared for redistribution. ASSOCIATIVE is the wrong key: the literature disagrees when studies are grouped by whether a task requires associations, because patients often pass low-resolution associative tests Quotation withheld because this record is not cleared for redistribution. HIGH-RESOLUTION is the proposed replacement key, and the paper's whole evidence is that regrouping the same studies by it removes the contradiction Quotation withheld because this record is not cleared for redistribution. a test of associative memory is not a process-pure measure of recollection, because familiarity and implicit memory can carry simple associations too Quotation withheld because this record is not cleared for redistribution. the same task can require the hippocampus or not depending on the STRATEGY the participant adopts - feature search needs no high-resolution representation, configural matching does Quotation withheld because this record is not cleared for redistribution. whether a study finds an impairment at all can depend on the participants' response criterion, which binary same/different designs never measure Quotation withheld because this record is not cleared for redistribution. the deficit is in PRECISION, not capacity: patients remember roughly which colour, and not exactly which - so an accuracy score conflates two things the account separates Quotation withheld because this record is not cleared for redistribution. claim_type: critique a second reading DIFFERING from the recorded value. Recorded: mechanism — 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: review a second reading reaching the same value. Recorded: review — 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. phenomenon: hippocampal patients failing and passing structurally similar associative tasks in different studies a second reading DIFFERING from the recorded value. Recorded: hippocampal high-resolution binding — 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 → dual-process models of recollection and familiarity (Atkinson & Juola, 1974; Yonelinas 1994) Quotation withheld because this record is not cleared for redistribution. extends → relational/binding models of hippocampal function (Cohen, Poldrack & Eichenbaum, 1997; Eichenbaum, Ranganath & Yonelinas, 2007) Quotation withheld because this record is not cleared for redistribution. contradicts → traditional memory-systems models in which the MTL is specialised for long-term memory (Squire, 1994; Schacter & Tulving, 1984; Tulving, 1982) differs in: THE AXIS IS THE PRINCIPLE OF DIVISION. The models contradicted 'assume that the MTL is specialized for long-term memory whereas other cortical systems are responsible for working memory and perception' - a division by PROCESS. The current approach divides by representational resolution instead, so the same MTL findings are partitioned differently rather than disputed. Quotation withheld because this record is not cleared for redistribution. contradicts → accounts of a selective hippocampal role in spatial processing (Lee et al. 2005; O'Keefe & Nadel, 1978) Quotation withheld because this record is not cleared for redistribution. contradicts → the claim that the hippocampus is required for any relational information or association (Konkel, Warren, Duff, Tranel, & Cohen, 2008) differs in: the RESOLUTION of the association, not its existence: NOT THAT THE HIPPOCAMPUS IS NECESSARY FOR ANY TYPE OF RELATIONAL INFORMATION OR ASSOCIATION, BUT RATHER THAT IT WILL BE PARTICULARLY CRITICAL FOR HIGH-RESOLUTION ASSOCIATIONS Quotation withheld because this record is not cleared for redistribution. contradicts → the account in which the hippocampus is necessary for relational or associative information as such differs in: the RESOLUTION of the association: A SIMPLE RELATIONAL OR ASSOCIATIVE ACCOUNT IN AND OF ITSELF is not what is claimed - the division is high-resolution against low Quotation withheld because this record is not cleared for redistribution. contradicts → memory systems models in which the medial temporal lobe is specialised for long-term memory while other systems handle working memory and perception differs in: what the hippocampus is a component OF: THE CURRENT THEORY SUGGESTS THAT THE HIPPOCAMPUS IS NOT JUST A COMPONENT OF A SPECIALIZED LONG-TERM MEMORY SYSTEM, BUT RATHER IT SUPPORTS HIGH-RESOLUTION BINDINGS THAT ARE IMPORTANT FOR LONG-TERM MEMORY AS WELL AS WORKING MEMORY AND PERCEPTION - the older models divide by TIMESCALE, this one by RESOLUTION Quotation withheld because this record is not cleared for redistribution. contradicts → accounts assigning the hippocampus a selective role in spatial processing differs in: whether the material is spatial: ALTHOUGH SPACE/SCENE INFORMATION OFTEN DOES RELY ON COMPLEX HIGH-RESOLUTION ASSOCIATIVE INFORMATION... THE ROLE OF THE HIPPOCAMPUS IS NOT EXPECTED TO BE LIMITED TO TASKS THAT INVOLVE SPATIAL MATERIALS - space is a special case of resolution, not the criterion Quotation withheld because this record is not cleared for redistribution. reconciles → studies reporting hippocampal deficits in associative working memory and studies reporting none differs in: which binding tasks: IT IS NOT SIMPLY THAT THE HIPPOCAMPUS IS NECESSARY FOR WORKING MEMORY TASKS THAT REQUIRE BINDING OR ASSOCIATIVE INFORMATION BECAUSE NOT ALL TASKS THAT REQUIRE BINDING ARE DISRUPTED BY HIPPOCAMPAL DAMAGE Quotation withheld because this record is not cleared for redistribution. builds-on → dual-process models of recollection and familiarity Quotation withheld because this record is not cleared for redistribution. supports → the Complementary Learning Systems model's account of pattern separation in dentate gyrus and CA3 Quotation withheld because this record is not cleared for redistribution. high-resolution —contrasted-with→ low-resolution high-resolution low-resolution strength of the relation: 1 · note: THE PROPOSED JOIN KEY. Regrouping the same patient studies on this distinction is the paper's entire evidence Quotation withheld because this record is not cleared for redistribution. recollection —contrasted-with→ familiarity recollection familiarity strength of the relation: 1 · note: CRITERION: whether specific qualitative details come back, or only a sense of having seen it before Quotation withheld because this record is not cleared for redistribution. state-based —contrasted-with→ strength-based state-based strength-based strength of the relation: 1 · note: CRITERION: whether the observer can say WHAT differed, or only that something did Quotation withheld because this record is not cleared for redistribution. feature search —contrasted-with→ configural feature search configural differs in: THE AXIS IS WHETHER ONE FEATURE SUFFICES. 'in the perceptual matching task, the materials have been such that Quotation withheld [configuration].' The axis is a property of THE MATERIALS, not of the subjects or the task name - which is why a null on such a task is not evidence about configural processing. · strength of the relation: 1 · note: STRATEGY, NOT TASK: the same materials engage the hippocampus or not depending on how the participant chooses to solve them Quotation withheld because this record is not cleared for redistribution. precision —distinct-from→ capacity precision capacity differs in: THE AXIS IS PRECISION AGAINST CAPACITY. 'Quotation withheld' Both are ways of being worse at a memory task and a summary score cannot tell them apart - so a deficit reported without this distinction is compatible with two different accounts of what the hippocampus does. Cowan2008 in this corpus reports the same literature unable to decide whether its limit is time or capacity; here a third quantity is separated from capacity by measurement. · strength of the relation: 1 · note: the patients keep the second and lose the first, which an overall accuracy score would merge Quotation withheld because this record is not cleared for redistribution. color wheel —operationalised-as→ precision color wheel precision strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its span and kept, but a color-wheel task measured precision; operationalised-as is loose. Quotation withheld because this record is not cleared for redistribution. pattern separation —part-of→ CLS pattern separation CLS strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the CLS model 'assumes' the pattern-separation abilities of DG and CA3; part-of is loose. Quotation withheld because this record is not cleared for redistribution. perceptual matching —causes→ feature search perceptual matching feature search strength of the relation: 0.8 · note: the task's materials permitted a single-feature solution, which is the proposed reason its results conflict Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the matching task's materials ALLOW a feature-search strategy; causes is loose. Quotation withheld because this record is not cleared for redistribution. digit span —distinct-from→ high-resolution digit span high-resolution strength of the relation: 1 · note: so its long-standing preservation in amnesia is a PREDICTION of the account rather than an anomaly for it Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but digit span requires 'neither associative nor high-resolution information'; distinct-from is loose. Quotation withheld because this record is not cleared for redistribution. working memory —contrasted-with→ long-term memory working memory long-term memory differs in: THE AXIS IS WHETHER THE HIPPOCAMPUS IS REQUIRED FOR A SIMPLE ASSOCIATION. 'Quotation withheld.' The same associative content is involved on both sides; what changes is whether the structure is needed to hold it - a claim about the memory system rather than about the material. · strength of the relation: 1 · note: the same resolution demand has different consequences in each, because over short delays other regions can carry the binding Quotation withheld because this record is not cleared for redistribution. receiver operating characteristic (ROC) analysis receiver operating characteristic (ROC) analysis Quotation withheld because this record is not cleared for redistribution. signal detection model fit (unequal variance model) signal detection model fit (unequal variance model) Quotation withheld because this record is not cleared for redistribution. receiver operating characteristics receiver operating characteristics Quotation withheld because this record is not cleared for redistribution. Reviewing the patient literature, hippocampal damage impairs perception and working memory precisely on tasks demanding complex high-resolution bindings. Quotation withheld because this record is not cleared for redistribution. Hippocampal damage is much less likely to impair tasks that need only low-resolution or simple associations. Quotation withheld because this record is not cleared for redistribution. In working memory, hippocampal damage reduced the precision of the memories rather than the number of items retained. Quotation withheld because this record is not cleared for redistribution. In a same/different scene task the patients' perceptual deficit was confined to strength-based (relational match) responses, while state-based responses using local feature detail were intact. Quotation withheld because this record is not cleared for redistribution. In healthy participants, hippocampal fMRI activation scaled linearly with strength-based perceptual confidence and showed no extra rise for state-based responses. Quotation withheld because this record is not cleared for redistribution. Merely requiring an association does not predict hippocampal involvement, since some object-colour associative working memory tasks are performed normally after hippocampal damage. Quotation withheld because this record is not cleared for redistribution. Hippocampal involvement is not restricted to complex scene material: precise colour memory for coloured squares also depends on the hippocampus. Quotation withheld because this record is not cleared for redistribution. The threshold of complexity at which the hippocampus becomes necessary differs across domains: simple associations need it in long-term memory but not in working memory. Quotation withheld because this record is not cleared for redistribution. In long-term memory, selective hippocampal damage removes recollection while leaving familiarity-based recognition intact. Quotation withheld because this record is not cleared for redistribution. At a one-second delay hippocampal patients failed on object-location pairings while remaining normal for the objects and the locations on their own. Quotation withheld because this record is not cleared for redistribution. THE LITERATURE CONTRADICTS ITSELF: some studies find hippocampal deficits on associative working memory tasks and others using similar procedures find none Quotation withheld because this record is not cleared for redistribution. Olsen's object-location task showed a deficit at a 1 second delay for pairings but not for objects or locations alone Quotation withheld because this record is not cleared for redistribution. patient Jon performed normally on colour, shape, colour-shape and object-spoken-name working memory Quotation withheld because this record is not cleared for redistribution. the early-damage objection to that case was answered by patients damaged later in life showing the same preservation Quotation withheld because this record is not cleared for redistribution. THE OLD KEY FAILS: requiring binding is not sufficient to predict a hippocampal deficit, because not all binding tasks are disrupted Quotation withheld because this record is not cleared for redistribution. the tasks that found nothing may not have drawn heavily on high-resolution binding; studies requiring precise object locations do show a hippocampal role Quotation withheld because this record is not cleared for redistribution. CEILING EXCLUDED EXPLICITLY: preservation held across set sizes and therefore across a range of performance Quotation withheld because this record is not cleared for redistribution. the tasks were designed to be equally hard for controls, so patient-control differences could not reflect difficulty or ceiling Quotation withheld because this record is not cleared for redistribution. A CITED STUDY FLAGGED: one supporting result is reported with the note that no statistical comparisons were provided Quotation withheld because this record is not cleared for redistribution. patients' location errors were small displacements rather than random, so the memory was present but less precise Quotation withheld because this record is not cleared for redistribution. PRECISION NOT CAPACITY: patients were no less likely to remember a colour, but when they did their memory was less precise Quotation withheld because this record is not cleared for redistribution. and the effect appears with simple coloured squares, so the hippocampal role is not limited to complex scenes Quotation withheld because this record is not cleared for redistribution. in oddity judgement hippocampal patients were impaired for scenes but not objects, faces or colours Quotation withheld because this record is not cleared for redistribution. perceptual matching results conflict: some studies find scene deficits and others find normal performance on the same task Quotation withheld because this record is not cleared for redistribution. THE PROPOSED RESOLUTION IS STRATEGY: if the materials allow a single-feature solution the hippocampus is not needed, if they force configural matching it is Quotation withheld because this record is not cleared for redistribution. perceptual judgements split into a discrete state with access to specific details and a graded strength of overall match Quotation withheld because this record is not cleared for redistribution. the patients' deficit was confined to the strength-based half and left the state-based half untouched Quotation withheld because this record is not cleared for redistribution. and hippocampal activation in healthy imaging tracked strength-based perception linearly without rising for state-based Quotation withheld because this record is not cleared for redistribution. AN ARTEFACT THAT COULD DECIDE A LITERATURE: with a strict response criterion the same data would show no impairment, with a liberal one a significant impairment Quotation withheld because this record is not cleared for redistribution. the shortcoming named in earlier perception work is that it measured perception as a unitary phenomenon Quotation withheld because this record is not cleared for redistribution. the account is distinguished from a relational one: patients often pass associative tests needing only low-resolution associations Quotation withheld because this record is not cleared for redistribution. and from memory-systems models that confine the medial temporal lobe to long-term memory Quotation withheld because this record is not cleared for redistribution. and from spatial accounts: the role of the hippocampus is not expected to be limited to tasks that involve spatial materials Quotation withheld because this record is not cleared for redistribution. the allocentric-versus-egocentric debate is recast as a question about how many bindings a task demands Quotation withheld because this record is not cleared for redistribution. the hippocampus is proposed to be unique over long delays; as the delay shortens, other cortical processes can support performance, at least temporarily Quotation withheld because this record is not cleared for redistribution. digit span requires neither association nor resolution, so its preservation in amnesia is predicted rather than anomalous Quotation withheld because this record is not cleared for redistribution. A PROGRAMME FOR RE-EXAMINING OLD NULLS: earlier studies that found no deficit are worth rerunning with higher resolution demands Quotation withheld because this record is not cleared for redistribution. the unifying principle proposed is the demand for representing, maintaining and retrieving complex high-resolution bindings Quotation withheld because this record is not cleared for redistribution. familiarity-based recognition familiarity-based recognition Quotation withheld because this record is not cleared for redistribution. familiarity familiarity Quotation withheld because this record is not cleared for redistribution. receiver operating characteristics receiver operating characteristics Quotation withheld because this record is not cleared for redistribution. strength-based perception strength-based perception Quotation withheld because this record is not cleared for redistribution. recollection recollection Quotation withheld because this record is not cleared for redistribution. state-based state-based Quotation withheld because this record is not cleared for redistribution. CLS CLS Quotation withheld because this record is not cleared for redistribution. high-resolution high-resolution Quotation withheld because this record is not cleared for redistribution. high-resolution high-resolution Quotation withheld because this record is not cleared for redistribution. precision precision Quotation withheld because this record is not cleared for redistribution. response criterion response criterion Quotation withheld because this record is not cleared for redistribution. 2, 3 and 4 items — the small set sizes at which hippocampal patients were already impaired on precise object placement Quotation withheld because this record is not cleared for redistribution. 180 ms to 1.5 seconds — range of stimulus presentation rates over which the same/different perception results replicated Quotation withheld because this record is not cleared for redistribution. 4 colored squares for 250 ms, followed by a 1 s blank screen — stimulus timing of the colour-wheel working memory task used with hippocampal patients Quotation withheld because this record is not cleared for redistribution. 1 to 7 objects — set sizes in the precise-placement task where patients failed from two items up 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/j.bbr.2013.05.030.