Time cells in the hippocampus: a new dimension for mapping memories

Time cells in the hippocampus: a new dimension for mapping memories Eichenbaum

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.

Time cells in the hippocampus: a new dimension for mapping memories 2014 Nature Reviews Neuroscience 10.1038/nrn3827 Eichenbaum2014-time-cells-in-the-hippocampus-a-new-dimensio Eichenbaum2014-time-cells-in-the-hippocampus-a-new-dimensio.txt, the project's staged copy 4fe50e96d5765783e01322690f189dd9232f5937 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 hippocampal neurons code elapsed time as a dimension in its own right, and how that coding relates to spatial coding Quotation withheld because this record is not cleared for redistribution. the firing properties of time cells parallel those of place cells point for point, so time is an additional dimension integrated with spatial mapping rather than a separate system Quotation withheld because this record is not cleared for redistribution. existence Quotation withheld because this record is not cleared for redistribution. hippocampal time cells Quotation withheld because this record is not cleared for redistribution. review review Quotation withheld because this record is not cleared for redistribution. mechanisms Slowly evolving cortical temporal-context signals select among preconfigured intra-hippocampal firing chains, so that a given chain comes to represent a particular temporally organised memory Quotation withheld because this record is not cleared for redistribution. mechanisms Cortical areas of the 'where' stream deliver temporal-context signals to the hippocampus, which then parses that continuous context into discrete time fields Quotation withheld because this record is not cleared for redistribution. mechanisms Repeated sequential activation strengthens connections so each place cell is driven by an earlier-firing cell, producing the backward expansion of place fields Quotation withheld because this record is not cleared for redistribution. temporal information generated in the cerebral cortex and other brain areas may provide signals to time cells, or the phenomenon may arise from firing chains generated locally within the hippocampal circuitry Quotation withheld because this record is not cleared for redistribution. the chaining account is supported by place fields expanding backwards from their centre with repeated runs, as if each cell were driven by an earlier-firing one Quotation withheld because this record is not cleared for redistribution. the proposed resolution is a slowly evolving temporal context selecting among preconfigured firing chains, rather than either alone Quotation withheld because this record is not cleared for redistribution. prefrontal and parietal areas of the where stream turn out to be involved in the perception of time and project strongly to the hippocampus Quotation withheld because this record is not cleared for redistribution. The origin of hippocampal timing signals - external input versus internally generated firing chains - is left unresolved. what this limits: any mechanistic reading of time cells: IT REMAINS A CENTRAL QUESTION WHETHER TEMPORAL INFORMATION ... ORIGINATES FROM ITS INPUTS OR IS GENERATED INTERNALLY Quotation withheld because this record is not cleared for redistribution. The evidence reviewed is almost entirely restricted to hippocampal area CA1. what this limits: the anatomical reach of the whole review: STUDIES OF TIME CELLS HAVE FOCUSED ON HIPPOCAMPAL AREA CA1 BECAUSE THIS AREA SEEMS TO BE THE MOST IMPORTANT FOR TEMPORAL ORGANIZATION Quotation withheld because this record is not cleared for redistribution. The temporal/spatial distinction may be superficial if both reduce to association by proximity. what this limits: the time/space distinction the review is built on: IT REMAINS ENTIRELY POSSIBLE THAT THE COMMON THREAD IN BOTH DIMENSIONS IS 'ASSOCIATION BY PROXIMITY' IN SPACE OR TIME Quotation withheld because this record is not cleared for redistribution. whether the temporal signal comes from the inputs, is generated internally, or both, is left as the central open question duplicate of: limitations[0] · what this limits: the same scope as limitations[0] - inputs, internal generation, or a combination, all still open Quotation withheld because this record is not cleared for redistribution. one study reports new temporal patterns emerging with learning rather than existing chains being co-opted, against the preconfigured account what this limits: the co-opted-chains account specifically: one study Quotation withheld Quotation withheld because this record is not cleared for redistribution. the human recordings show slowly changing patterns rather than the punctate time-cell activations seen in rodents, so the cross-species claim rests on a phenomenon of different form what this limits: the human-to-rodent join: the human patterns MORE CLOSELY RESEMBLE THE GRADUALLY EVOLVING TIME SIGNAL ... THAN THE SEQUENCES OF PUNCTATE TIME-CELL ACTIVATIONS in rodents, and are still read as the same phenomenon Quotation withheld because this record is not cleared for redistribution. studies of time cells have focused on hippocampal area CA1, with some evidence for CA2 but not CA3 what this limits: the same scope as limitations[1], with what is known outside CA1: THERE IS SOME EVIDENCE THAT AREA CA2 (BUT NOT CA3) COULD ALSO BE IMPORTANT FOR TEMPORAL PROCESSING Quotation withheld because this record is not cleared for redistribution. the author's closing possibility is that time cells and place cells are both reflections of association by proximity, which would dissolve the distinction the review is built on what this limits: the same scope as limitations[2], with the consequence spelled out: PLACE CELLS AND TIME CELLS ARE SIMPLY REFLECTIONS OF A BASIC FUNCTION OF THE HIPPOCAMPUS Quotation withheld because this record is not cleared for redistribution. the cue-and-structure vocabulary carried over term by term: temporal cues answering to spatial cues, temporal structure to spatial structure Quotation withheld because this record is not cleared for redistribution. scalar timing, imported as the criterion a neural code must satisfy to count as a code for time Quotation withheld because this record is not cleared for redistribution. time cell and place cell are two names for ONE population - the same neurons, distinguished by which dimension the experiment lets you measure, not by which cells they are Quotation withheld because this record is not cleared for redistribution. the parallel is term for term, not loose analogy: temporal cues answer to spatial cues, temporal structure to spatial structure, re-timing to re-mapping, time fields to place fields Quotation withheld because this record is not cleared for redistribution. the name 'time cell' is a coinage of one particular study, made on the strength of an analogy to place cells - because they fire at moments as place cells fire at locations - rather than on a mechanism Quotation withheld because this record is not cleared for redistribution. scalar coding - later cells firing for longer - is not decoration but the property that adjudicates between the two mechanisms, since uniform-duration chains cannot produce it Quotation withheld because this record is not cleared for redistribution. temporal coding in humans means gradually changing population patterns, while in rodents it means punctate successive activations; the review says so while treating both as the same finding Quotation withheld because this record is not cleared for redistribution. judging an interval and remembering temporal organisation are separated, with the hippocampus central only to the second - so 'timing' names two roles with different anatomy Quotation withheld because this record is not cleared for redistribution. claim_type: existence a second reading reaching the same value. Recorded: existence — 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 neurons firing at successive specific moments within a temporally structured period, filling it out a second reading DIFFERING from the recorded value. Recorded: hippocampal time cells — 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. contradicts → the view that the fundamental function of the hippocampus is spatial mapping and navigation (the paper's reference 75) Quotation withheld because this record is not cleared for redistribution. builds-on → the scalar temporal context model (the paper's reference 69) Quotation withheld because this record is not cleared for redistribution. contradicts → the view that the fundamental function of the hippocampus is spatial mapping and navigation Quotation withheld because this record is not cleared for redistribution. critiques → the strict chaining model in which time cells are a single fixed sequence of cell activations Quotation withheld because this record is not cleared for redistribution. builds-on → the temporal context model of hippocampal coding, in its scalar variant Quotation withheld because this record is not cleared for redistribution. qualifies → the original reports of hippocampal temporal coding, criticised as confounded by location and distance travelled differs in: THE AXIS IS A SPATIAL CONFOUND. The original temporal-coding reports 'were initially criticized as being Quotation withheld across time and of distance travelled over time, so that the apparent temporal coding could reflect spatial rathe[r than temporal] coding'. The finding is not denied - an alternative explanation for it is offered, on one named variable. Quotation withheld because this record is not cleared for redistribution. extends → the place-cell account of hippocampal function, by adding a temporal dimension carried by the same cells Quotation withheld because this record is not cleared for redistribution. time cells —same-as→ place cells time cells place cells strength of the relation: 1 · note: THE STRONGEST SAME-AS IN THE CORPUS: not two populations but one, given two names by two literatures. No lexical join could find this Quotation withheld because this record is not cleared for redistribution. scalar coding —operationalised-as→ time-fields scalar coding time-fields strength of the relation: 1 · note: scalar time is read off the width of the field, not off any clock Audit 2026-09-28: DOUBTFUL - read against its span and kept, but larger time-fields later in the delay is scalar coding; operationalised-as is loose. Quotation withheld because this record is not cleared for redistribution. firing chains —contrasted-with→ temporal context firing chains temporal context strength of the relation: 1 · note: CRITERION: uniform-duration chains cannot produce later cells that fire longer, and scalar firing is what the temporal context model predicts Quotation withheld because this record is not cleared for redistribution. temporal context —contrasted-with→ chaining model temporal context chaining model strength of the relation: 1 · note: external source against internal generation - the two candidate origins the review is organised around Quotation withheld because this record is not cleared for redistribution. temporal context —causes→ firing chains temporal context firing chains strength of the relation: 0.8 · note: the proposed resolution: context SELECTS among chains rather than either mechanism acting alone. A proposal, hedged as 'perhaps the most likely' - weighted accordingly Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but 'perhaps the most likely' mechanism involves temporal context selecting firing chains; causes is hedged. Quotation withheld because this record is not cleared for redistribution. elapsed time —distinct-from→ distance travelled elapsed time distance travelled strength of the relation: 1 · note: DECOUPLED BY DESIGN: treadmill speed varied randomly per trial so the two could be separated at all Quotation withheld because this record is not cleared for redistribution. distance travelled —contrasted-with→ elapsed time distance travelled elapsed time strength of the relation: 1 · note: AND THE CONFOUND SURVIVED: most cells encoded BOTH, to differing degrees - decoupling vindicated the variable without eliminating the rival Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but most neurons encoded BOTH time and distance; contrasted-with is loose. Quotation withheld because this record is not cleared for redistribution. grid cells —part-of→ entorhinal cortex grid cells entorhinal cortex strength of the relation: 1 · note: and they too fire at particular moments, so the same dual coding appears one synapse upstream Quotation withheld because this record is not cleared for redistribution. CA1 —contrasted-with→ CA2 CA1 CA2 strength of the relation: 0.7 · note: SCOPE: nearly all the evidence is CA1; CA2 has some support and CA3 explicitly none Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but CA1 is where time cells were studied and CA2 'could also be important'; contrasted-with is loose. Quotation withheld because this record is not cleared for redistribution. trace eye-blink conditioning —distinct-from→ firing chains trace eye-blink conditioning firing chains differs in: THE AXIS IS WHETHER LEARNING CREATES THE PATTERN OR RECRUITS ONE. 'one study examining trace eye-blink conditioning Quotation withheld.' Both accounts predict temporally structured firing after learning and differ only on where the structure came from - which is invisible in the trained animal and visible only across learning. · strength of the relation: 0.8 · note: the dissenting paradigm: new patterns EMERGED with learning rather than existing chains being co-opted Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but trace eye-blink conditioning produced NEW temporal patterns rather than co-opting firing chains; distinct-from is loose. Quotation withheld because this record is not cleared for redistribution. Hippocampal CA1 principal cells encode the temporal structure of remembered experience independently of the animal's location and movement. Quotation withheld because this record is not cleared for redistribution. Time cells and place cells are not distinct populations — the same hippocampal neurons carry both codes. Quotation withheld because this record is not cleared for redistribution. Time-cell firing sequences emerge with learning, are specific to particular memories, and predict whether a subsequent memory judgement will be correct. Quotation withheld because this record is not cleared for redistribution. Several findings suggest a strict chaining model is unlikely to be the mechanism for time cells. Quotation withheld because this record is not cleared for redistribution. This challenges the view that spatial mapping and navigation is the hippocampus's fundamental function. Quotation withheld because this record is not cleared for redistribution. the original temporal-coding reports were criticised as confounded by the animal's changing location and by distance travelled over time Quotation withheld because this record is not cleared for redistribution. later studies addressed this by tightly controlling the animal's location and movement Quotation withheld because this record is not cleared for redistribution. FIRST RUNG: cells with reliable place fields were simply excluded from the analysis Quotation withheld because this record is not cleared for redistribution. SECOND RUNG: a general linear model separated location, speed and head direction from elapsed time, leaving an independent contribution of time Quotation withheld because this record is not cleared for redistribution. THIRD RUNG, THE DECOUPLING: treadmill speed was randomised per trial so that elapsed time and distance travelled could be disentangled Quotation withheld because this record is not cleared for redistribution. over 80 percent of active neurons were modulated by elapsed time, position mattered little, but distance was also strongly encoded and most cells carried both Quotation withheld because this record is not cleared for redistribution. FOURTH RUNG: head-fixed preparations removed location and movement altogether Quotation withheld because this record is not cleared for redistribution. the conclusion the ladder licenses: CA1 principal cells encode temporal structure independently of location and movement Quotation withheld because this record is not cleared for redistribution. AN EARLY DISSOCIATION: spatial representations differed across trials too, but equally on correct and error trials, so they did not predict memory accuracy Quotation withheld because this record is not cleared for redistribution. the firing patterns of many neurons re-timed when the delay was increased, even though the behaviour and locations of the animal during the initial period did not change Quotation withheld because this record is not cleared for redistribution. cells firing later in the delay were active for longer, a scalar coding of time paralleling human and animal time judgements WHOSE RESULT THIS IS: the authors 18 observed - 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 IDENTITY CLAIM: time cells are the same neurons as place cells, and in freely moving animals fire only when the animal is in a particular location Quotation withheld because this record is not cleared for redistribution. the same cells that fired at particular moments on the treadmill fired at particular places on the maze outside it Quotation withheld because this record is not cleared for redistribution. the reorganisation of time cells when a temporal cue changes parallels the reorganisation of place cells when spatial cues change Quotation withheld because this record is not cleared for redistribution. temporal cues and structure control time cells just as spatial cues and structure control place cells Quotation withheld because this record is not cleared for redistribution. the proposal that follows is to treat hippocampal principal neurons as multidimensional rather than as either kind of cell Quotation withheld because this record is not cleared for redistribution. temporal coding is tied to memory three ways: it develops with learning, it is memory-specific, and it predicts judgement accuracy Quotation withheld because this record is not cleared for redistribution. A DOUBLE GRADIENT IN MONKEYS: time coding falls and object-identity coding rises going from entorhinal to perirhinal to inferotemporal cortex Quotation withheld because this record is not cleared for redistribution. QUALIFICATION ON THE CROSS-SPECIES CLAIM: the human patterns are slowly changing rather than punctate, closer to an earlier rat result than to the rodent time cells Quotation withheld because this record is not cleared for redistribution. place-cell sequences replay in compressed form during sleep and quiet wakefulness, so spatial coding is recapitulated temporally Quotation withheld because this record is not cleared for redistribution. when spatial and sequence strategies compete, the firing sequences follow whichever strategy the animal expressed on that probe trial Quotation withheld because this record is not cleared for redistribution. FOUR ARGUMENTS AGAINST STRICT CHAINING, first: the cells changed their firing patterns even during the initial part of the delay, rather than simply adding cell activations during the extended segment Quotation withheld because this record is not cleared for redistribution. second: strict chains cannot split to encode overlapping memories that share identical events Quotation withheld because this record is not cleared for redistribution. third: scalar firing is inconsistent with uniform-duration chain activations and consistent with the scalar temporal context model Quotation withheld because this record is not cleared for redistribution. fourth: compressed replay would need a mechanism that can decompress back to the original intervals Quotation withheld because this record is not cleared for redistribution. on the other side, firing chains appear to be preconfigured and then co-opted to organise events Quotation withheld because this record is not cleared for redistribution. DISSENT PRESERVED: one trace eye-blink study found new temporal patterns emerging with learning rather than existing chains being used WHOSE RESULT THIS IS: one study examining trace eye-blink conditioning reported - 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 proposed synthesis is temporal context dynamics selecting specific firing chains Quotation withheld because this record is not cleared for redistribution. grid cells in medial entorhinal cortex also fire at particular moments during treadmill running Quotation withheld because this record is not cleared for redistribution. judging intervals and remembering temporal organisation are separated, with the hippocampus secondary to the first and central to the second Quotation withheld because this record is not cleared for redistribution. the review's headline consequence: this challenges spatial mapping and navigation as the fundamental function of the hippocampus Quotation withheld because this record is not cleared for redistribution. THE AUTHOR'S CLOSING DEFLATION: it remains entirely possible that both dimensions reduce to association by proximity, making time cells and place cells reflections of one associative function Quotation withheld because this record is not cleared for redistribution. time cells time cells Quotation withheld because this record is not cleared for redistribution. place cells place cells Quotation withheld because this record is not cleared for redistribution. re-timing re-timing Quotation withheld because this record is not cleared for redistribution. temporal context model temporal context model Quotation withheld because this record is not cleared for redistribution. chaining model chaining model Quotation withheld because this record is not cleared for redistribution. place cells place cells Quotation withheld because this record is not cleared for redistribution. time cells time cells Quotation withheld because this record is not cleared for redistribution. scalar coding scalar coding WHOSE RESULT THIS IS: the authors 18 observed - 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. replay replay Quotation withheld because this record is not cleared for redistribution. chaining model chaining model Quotation withheld because this record is not cleared for redistribution. over 80% — proportion of neurons active during treadmill running whose activity was modulated by elapsed time, in a study that dissociated time from distance travelled Quotation withheld because this record is not cleared for redistribution. approximately 30% — proportion of hippocampal cells coding successive moments of the delay in head-fixed rats doing a delayed matching-to-sample task Quotation withheld because this record is not cleared for redistribution. about 100 ms — the compressed duration within which seconds-long place-cell sequences are replayed during quiet periods Quotation withheld because this record is not cleared for redistribution. over 80% — neurons active during treadmill running whose activity was modulated by elapsed time 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. elapsed time elapsed time Quotation withheld because this record is not cleared for redistribution. distance travelled distance travelled 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/10.1038/nrn3827.