The what, where and how of delay activity

The what, where and how of delay activity Sreenivasan D’Esposito

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 what, where and how of delay activity 2019 Nature Reviews Neuroscience 10.1038/s41583-019-0176-7 Sreenivasan2019-the-what-where-and-how-of-delay-activity Sreenivasan2019-the-what-where-and-how-of-delay-activity.txt, the project's staged copy 663f4b2ad6cd7e357eea333daec4d8e193c277a8 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 DELAY ACTIVITY - NEURAL ACTIVITY THAT PERSISTS DURING THE RETENTION INTERVAL OF DELAY TASKS, reconsidered as to WHAT IT IS, WHERE IN THE BRAIN IT IS FOUND, WHAT ROLES IT SERVES AND HOW IT MAY BE GENERATED Quotation withheld because this record is not cleared for redistribution. what the object even is, once the methods have changed - METHODOLOGICAL AND ANALYTICAL ADVANCES HAVE ALLOWED RESEARCHERS TO UNCOVER PREVIOUSLY UNOBSERVED FORMS OF DELAY ACTIVITY ACROSS THE ENTIRE BRAIN Quotation withheld because this record is not cleared for redistribution. an expanded definition - ANY TASK-RELATED CHANGE IN NEURAL ACTIVITY THAT SPANS THE INTERVAL BETWEEN A STIMULUS AND THE BEHAVIOURAL RESPONSE IN A WM DELAY TASK - and an account of what it costs Quotation withheld because this record is not cleared for redistribution. critique Quotation withheld because this record is not cleared for redistribution. delay activity in working memory Quotation withheld because this record is not cleared for redistribution. review review Quotation withheld because this record is not cleared for redistribution. mechanisms Structured recurrent excitation with tuned excitation–inhibition balance lets network spiking settle into a self-sustaining attractor state that keeps the memory alive without input Quotation withheld because this record is not cleared for redistribution. mechanisms Slow NMDA-receptor excitation stabilises attractor networks so that persistent spiking can be maintained over seconds Quotation withheld because this record is not cleared for redistribution. mechanisms Encoding-driven changes in synaptic weights turn the network into a matched filter, so the memory is stored silently and revealed when noisy input is fed through it Quotation withheld because this record is not cleared for redistribution. mechanisms Feedforward connections between similarly tuned cells relay brief bursts of spiking along a chain, storing information without any single cell firing continuously Quotation withheld because this record is not cleared for redistribution. mechanisms Long-range input (from thalamus) is required for local cortical circuits to sustain delay activity at all Quotation withheld because this record is not cleared for redistribution. the definition is widened deliberately and its price is named in the same sentence: THIS EXPANDED DEFINITION ALLOWS FOR A MORE INCLUSIVE AND COMPREHENSIVE EVALUATION OF DELAY ACTIVITY, BUT AT THE SAME TIME CALLS ATTENTION TO THE CHALLENGE OF COMPARING DATA ACROSS DIFFERENT EXPERIMENTAL METHODS, BRAIN REGIONS AND SPECIES Quotation withheld because this record is not cleared for redistribution. and the one design that escapes the problem: WE ADDRESS THIS ISSUE BY HIGHLIGHTING STUDIES THAT RECORD DELAY ACTIVITY ACROSS MULTIPLE BRAIN REGIONS IN THE SAME EXPERIMENT Quotation withheld because this record is not cleared for redistribution. a code that cannot be read from its parts: Quotation withheld AND INSTEAD ARE ENCODED IN THE COMBINED ACTIVITY OF SEVERAL NEURONS Quotation withheld because this record is not cleared for redistribution. DELAY ACTIVITY IS SEEN MORE GENERALLY IN CONTEXTS THAT REQUIRE AN ORGANISM TO LINK A SENSORY STIMULUS TO A DELAYED BEHAVIOUR, SUCH AS DURING TASKS OF SUSTAINED ATTENTION OR DECISION MAKING; HERE, HOWEVER, WE EXAMINE DELAY ACTIVITY IN TASKS EXPLICITLY TESTING WM, WHERE IT HAS BEEN CHARACTERIZED MOST EXTENSIVELY what this limits: the attribution of delay activity to working memory: DELAY ACTIVITY IS SEEN MORE GENERALLY IN CONTEXTS THAT REQUIRE AN ORGANISM TO LINK A SENSORY STIMULUS TO A DELAYED BEHAVIOUR, SUCH AS DURING TASKS OF SUSTAINED ATTENTION OR DECISION MAKING Quotation withheld because this record is not cleared for redistribution. The broad definition the review adopts makes comparison across methods, regions and species harder. what this limits: every comparison the review makes under its own definition: it ALLOWS FOR A MORE INCLUSIVE AND COMPREHENSIVE EVALUATION ... BUT AT THE SAME TIME CALLS ATTENTION TO THE CHALLENGE OF COMPARING DATA ACROSS DIFFERENT EXPERIMENTAL METHODS, BRAIN REGIONS AND SPECIES Quotation withheld because this record is not cleared for redistribution. AS THE RELATIONSHIP BETWEEN THE BIOPHYSICAL PROPERTIES OF NEURONS OR MICRO-SCALE NETWORKS AND THE EFP OR BOLD SIGNALS IS INCOMPLETELY UNDERSTOOD, BIOPHYSICAL MODELS ALMOST EXCLUSIVELY ATTEMPT TO DESCRIBE SPIKING ACTIVITY; ACCORDINGLY, WE FOCUS OUR DISCUSSION HERE ON DELAY SPIKING what this limits: the coverage of the biophysical modelling literature: AS THE RELATIONSHIP BETWEEN THE BIOPHYSICAL PROPERTIES OF NEURONS OR MICRO-SCALE NETWORKS AND THE EFP OR BOLD SIGNALS IS INCOMPLETELY UNDERSTOOD, BIOPHYSICAL MODELS ALMOST EXCLUSIVELY ATTEMPT TO DESCRIBE SPIKING Quotation withheld because this record is not cleared for redistribution. the expanded definition makes the literature harder to compare, not easier - THE CHALLENGE OF COMPARING DATA ACROSS DIFFERENT EXPERIMENTAL METHODS, BRAIN REGIONS AND SPECIES what this limits: pooling delay-activity findings across techniques without asking what each technique can see Quotation withheld because this record is not cleared for redistribution. DIFFERENCES IN TASKS, METHODOLOGY AND SPECIES IN THE ABOVE STUDIES POSE A SIGNIFICANT OBSTACLE TO UNCOVERING THE LOCATION AND FUNCTION OF DELAY ACTIVITY what this limits: the localisation and function questions the review is asking: DIFFERENCES IN TASKS, METHODOLOGY AND SPECIES IN THE ABOVE STUDIES POSE A SIGNIFICANT OBSTACLE TO UNCOVERING THE LOCATION AND FUNCTION OF DELAY ACTIVITY Quotation withheld because this record is not cleared for redistribution. each instrument sees a different thing - RECORDING DELAY ACTIVITY AT THESE DIFFERENT SCALES REQUIRES MULTIPLE TECHNIQUES, EACH OF WHICH IS SENSITIVE TO DIFFERENT ASPECTS OF THE UNDERLYING NEURONAL SIGNALS what this limits: treating a delay signal recorded one way as the same object as one recorded another way Quotation withheld because this record is not cleared for redistribution. and for one of those instruments the source is unknown - CHALLENGES IN LOCALIZING EEG ACTIVITY MAKE IT UNCLEAR WHERE IN THE BRAIN THESE SIGNALS ORIGINATE what this limits: the EEG evidence specifically: CHALLENGES IN LOCALIZING EEG ACTIVITY MAKE IT UNCLEAR WHERE IN THE BRAIN THESE SIGNALS ORIGINATE Quotation withheld because this record is not cleared for redistribution. the interpretation of the burst data is contested - Quotation withheld what this limits: the LFP burst evidence: Quotation withheld Quotation withheld because this record is not cleared for redistribution. and for one of those instruments the source is unknown - CHALLENGES IN LOCALIZING EEG ACTIVITY MAKE IT UNCLEAR WHERE IN THE BRAIN THESE SIGNALS ORIGINATE Quotation withheld because this record is not cleared for redistribution. the review restricted, where possible, to experiments that record several regions at once eliminates: differences between studies masquerading as differences between regions - WE ADDRESS THIS ISSUE BY HIGHLIGHTING STUDIES THAT RECORD DELAY ACTIVITY ACROSS MULTIPLE BRAIN REGIONS IN THE SAME EXPERIMENT · outcome (stated by the tagger, no quotation): partial Quotation withheld because this record is not cleared for redistribution. THE DEFINITION IS WIDENED IN THE SECOND PARAGRAPH, AND THE COST IS IN THE SAME SENTENCE: Quotation withheld - THIS EXPANDED DEFINITION ALLOWS FOR A MORE INCLUSIVE AND COMPREHENSIVE EVALUATION OF DELAY ACTIVITY, BUT AT THE SAME TIME CALLS ATTENTION TO THE CHALLENGE OF COMPARING DATA ACROSS DIFFERENT EXPERIMENTAL METHODS, BRAIN REGIONS AND SPECIES Quotation withheld because this record is not cleared for redistribution. and what it replaces is named: DISCUSSIONS OF DELAY ACTIVITY THAT TYPICALLY FOCUS ON INCREASES IN SPIKE RATE OR FUNCTIONAL MRI (fMRI) SIGNAL THAT PERSIST THROUGHOUT A WM DELAY. The older definition is a MEASUREMENT; the new one is a RELATION to the task Quotation withheld because this record is not cleared for redistribution. THE PHENOMENON GREW WHEN THE INSTRUMENTS CHANGED: ALTHOUGH INITIALLY THERE WAS A NARROW FOCUS ON SUSTAINED ACTIVATION IN A SMALL NUMBER OF BRAIN REGIONS, METHODOLOGICAL AND ANALYTICAL ADVANCES HAVE ALLOWED RESEARCHERS TO UNCOVER PREVIOUSLY UNOBSERVED FORMS OF DELAY ACTIVITY ACROSS THE ENTIRE BRAIN Quotation withheld because this record is not cleared for redistribution. and each instrument sees a different thing: RECORDING DELAY ACTIVITY AT THESE DIFFERENT SCALES REQUIRES MULTIPLE TECHNIQUES, EACH OF WHICH IS SENSITIVE TO DIFFERENT ASPECTS OF THE UNDERLYING NEURONAL SIGNALS Quotation withheld because this record is not cleared for redistribution. THE CLASSIC RESULT IS A PRODUCT OF THE AVERAGE: ONLY AFTER AVERAGING ACTIVITY OVER TRIALS DID THE CLASSIC PATTERNS OF SUSTAINED GAMMA AND BETA ENHANCEMENT EMERGE. The sustained pattern that named the phenomenon is not present in any single trial Quotation withheld because this record is not cleared for redistribution. and the underlying signal is disputed: Quotation withheld Quotation withheld because this record is not cleared for redistribution. THE OBSTACLE IS THE SAME THREE THINGS EVERY TIME: DIFFERENCES IN TASKS, METHODOLOGY AND SPECIES IN THE ABOVE STUDIES POSE A SIGNIFICANT OBSTACLE TO UNCOVERING THE LOCATION AND FUNCTION OF DELAY ACTIVITY Quotation withheld because this record is not cleared for redistribution. and the only clean design is named: WE ADDRESS THIS ISSUE BY HIGHLIGHTING STUDIES THAT RECORD DELAY ACTIVITY ACROSS MULTIPLE BRAIN REGIONS IN THE SAME EXPERIMENT - compare within one experiment, because between experiments the comparison is not available Quotation withheld because this record is not cleared for redistribution. A CODE THAT CANNOT BE READ FROM ITS PARTS: NONLINEAR MIXED SELECTIVITY ENABLES INDIVIDUAL NEURONS TO RESPOND TO COMBINATIONS OF TASK AND STIMULUS FEATURES SUCH THAT FEATURE COMBINATIONS CANNOT BE READ OUT FROM THE RESPONSES OF THE INDIVIDUAL NEURON ALONE AND INSTEAD ARE ENCODED IN THE COMBINED ACTIVITY OF SEVERAL NEURONS. A per-neuron tag cannot carry what the population encodes Quotation withheld because this record is not cleared for redistribution. and it is offered as a reconciliation of two positions: ATTEMPTS TO RECONCILE THESE PERSPECTIVES HAVE GIVEN RISE TO A MORE NUANCED VIEW, THAT lPFC DELAY ACTIVITY SIMULTANEOUSLY REPRESENTS MULTIPLE DIMENSIONS OF WM CONTENT AND GOALS Quotation withheld because this record is not cleared for redistribution. AND FOR ONE INSTRUMENT THE SOURCE OF THE SIGNAL IS UNKNOWN: CHALLENGES IN LOCALIZING EEG ACTIVITY MAKE IT UNCLEAR WHERE IN THE BRAIN THESE SIGNALS ORIGINATE - so a `where` claim from that instrument is not a claim about a place Quotation withheld because this record is not cleared for redistribution. the review's own framing of what it is doing: IN LIGHT OF THESE NEW FINDINGS, THIS REVIEW RECONSIDERS WHAT DELAY ACTIVITY IS, WHERE IN THE BRAIN IT IS FOUND, WHAT ROLES IT SERVES AND HOW IT MAY BE GENERATED Quotation withheld because this record is not cleared for redistribution. and it is offered as a reconciliation of two positions: ATTEMPTS TO RECONCILE THESE PERSPECTIVES HAVE GIVEN RISE TO A MORE NUANCED VIEW, THAT lPFC DELAY ACTIVITY SIMULTANEOUSLY REPRESENTS MULTIPLE DIMENSIONS OF WM CONTENT AND GOALS Quotation withheld because this record is not cleared for redistribution. AND FOR ONE INSTRUMENT THE SOURCE OF THE SIGNAL IS UNKNOWN: CHALLENGES IN LOCALIZING EEG ACTIVITY MAKE IT UNCLEAR WHERE IN THE BRAIN THESE SIGNALS ORIGINATE - so a `where` claim from that instrument is not a claim about a place Quotation withheld because this record is not cleared for redistribution. claim_type: critique a second reading reaching the same value. Recorded: critique — 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: review across scales and techniques, giving weight to studies that record in multiple regions within one experiment because that is the only way to compare 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: a fifty-year-old phenomenon whose extent grew when the instruments changed, not when the brain did a second reading DIFFERING from the recorded value. Recorded: delay activity in working memory — 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. critiques → the narrow definition of delay activity as sustained spike-rate or fMRI increase differs in: what counts as delay activity - ANY TASK-RELATED CHANGE IN NEURAL ACTIVITY THAT SPANS THE INTERVAL BETWEEN A STIMULUS AND THE BEHAVIOURAL RESPONSE Quotation withheld because this record is not cleared for redistribution. reconciles → the positions that lPFC delay activity codes content and that it codes goals differs in: the assumption that a neuron codes one thing - NONLINEAR MIXED SELECTIVITY ENABLES INDIVIDUAL NEURONS TO RESPOND TO COMBINATIONS OF TASK AND STIMULUS FEATURES Quotation withheld because this record is not cleared for redistribution. qualifies → the classic sustained-enhancement description of delay activity differs in: whether trials are averaged - ONLY AFTER AVERAGING ACTIVITY OVER TRIALS DID THE CLASSIC PATTERNS OF SUSTAINED GAMMA AND BETA ENHANCEMENT EMERGE Quotation withheld because this record is not cleared for redistribution. reconciles → the positions that lPFC delay activity codes content and that it codes goals differs in: the assumption that a neuron codes one thing - NONLINEAR MIXED SELECTIVITY ENABLES INDIVIDUAL NEURONS TO RESPOND TO COMBINATIONS OF TASK AND STIMULUS FEATURES Quotation withheld because this record is not cleared for redistribution. delay activity —part-of→ WM delay activity WM depends on: which definition is in force - the older one focuses on INCREASES IN SPIKE RATE OR FUNCTIONAL MRI (fMRI) SIGNAL THAT PERSIST THROUGHOUT A WM DELAY · strength of the relation: 1 · note: ANY TASK-RELATED CHANGE IN NEURAL ACTIVITY THAT SPANS THE INTERVAL BETWEEN A STIMULUS AND THE BEHAVIOURAL RESPONSE IN A WM DELAY TASK Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but delay activity is defined as activity spanning a WM delay; part-of is loose. Quotation withheld because this record is not cleared for redistribution. delay activity —operationalised-as→ fMRI delay activity fMRI strength of the relation: 0.5 — the record marks this weak · note: the definition being replaced: DISCUSSIONS OF DELAY ACTIVITY THAT TYPICALLY FOCUS ON INCREASES IN SPIKE RATE OR FUNCTIONAL MRI (fMRI) SIGNAL THAT PERSIST THROUGHOUT A WM DELAY Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the paper widens delay activity BEYOND spike-rate and fMRI increases; operationalised-as fMRI is loose. Quotation withheld because this record is not cleared for redistribution. gamma —contrasted-with→ beta gamma beta strength of the relation: 1 · note: beta bursts SEEMED TO INHIBIT GAMMA BURSTS AND REDUCED THE INFORMATION CONTENT OF SPIKING ACTIVITY Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span reports gamma bursts; the beta contrast is not in the shown text. Quotation withheld because this record is not cleared for redistribution. One lateral prefrontal study estimated that only 3% of delay-active neurons spike at a stable rate throughout the delay. Quotation withheld because this record is not cleared for redistribution. Quotation withheld Quotation withheld because this record is not cleared for redistribution. Optogenetically inhibiting thalamic delay activity disrupted delay activity in frontal regions. Quotation withheld because this record is not cleared for redistribution. Thalamic delay activity is not a copy of the cortical signal — it is less content-selective than frontal delay activity yet causally sets frontal selectivity. Quotation withheld because this record is not cleared for redistribution. Quotation withheld Quotation withheld because this record is not cleared for redistribution. Which region shows delay activity depends on the animal's strategy: a prospective motor strategy raised prefrontal delay activity while a retrospective sensory strategy raised it in posterior cortex. Quotation withheld because this record is not cleared for redistribution. In monkey lPFC, drift in population delay spiking predicted the size of the behavioural error. Quotation withheld because this record is not cleared for redistribution. Blocking prefrontal NMDA receptors eliminates delay spiking in non-human primates. Quotation withheld because this record is not cleared for redistribution. No single biophysical model reproduces every property of delay activity, and most empirical results are equally compatible with several model classes. Quotation withheld because this record is not cleared for redistribution. Quotation withheld Quotation withheld because this record is not cleared for redistribution. THIS EXPANDED DEFINITION ALLOWS FOR A MORE INCLUSIVE AND COMPREHENSIVE EVALUATION OF DELAY ACTIVITY, BUT AT THE SAME TIME CALLS ATTENTION TO THE CHALLENGE OF COMPARING DATA ACROSS DIFFERENT EXPERIMENTAL METHODS, BRAIN REGIONS AND SPECIES Quotation withheld because this record is not cleared for redistribution. METHODOLOGICAL AND ANALYTICAL ADVANCES HAVE ALLOWED RESEARCHERS TO UNCOVER PREVIOUSLY UNOBSERVED FORMS OF DELAY ACTIVITY ACROSS THE ENTIRE BRAIN Quotation withheld because this record is not cleared for redistribution. ONLY AFTER AVERAGING ACTIVITY OVER TRIALS DID THE CLASSIC PATTERNS OF SUSTAINED GAMMA AND BETA ENHANCEMENT EMERGE Quotation withheld because this record is not cleared for redistribution. DIFFERENCES IN TASKS, METHODOLOGY AND SPECIES IN THE ABOVE STUDIES POSE A SIGNIFICANT OBSTACLE TO UNCOVERING THE LOCATION AND FUNCTION OF DELAY ACTIVITY Quotation withheld because this record is not cleared for redistribution. FEATURE COMBINATIONS CANNOT BE READ OUT FROM THE RESPONSES OF THE INDIVIDUAL NEURON ALONE AND INSTEAD ARE ENCODED IN THE COMBINED ACTIVITY OF SEVERAL NEURONS Quotation withheld because this record is not cleared for redistribution. CHALLENGES IN LOCALIZING EEG ACTIVITY MAKE IT UNCLEAR WHERE IN THE BRAIN THESE SIGNALS ORIGINATE Quotation withheld because this record is not cleared for redistribution. Quotation withheld Quotation withheld because this record is not cleared for redistribution. delay spiking delay spiking Quotation withheld because this record is not cleared for redistribution. population coding population coding Quotation withheld because this record is not cleared for redistribution. bistability bistability Quotation withheld because this record is not cleared for redistribution. activity silent models activity silent models Quotation withheld because this record is not cleared for redistribution. nonlinear mixed selectivity nonlinear mixed selectivity Quotation withheld because this record is not cleared for redistribution. sensory recruitment model sensory recruitment model Quotation withheld because this record is not cleared for redistribution. ANY TASK-RELATED CHANGE IN NEURAL ACTIVITY THAT SPANS THE INTERVAL BETWEEN A STIMULUS AND THE BEHAVIOURAL RESPONSE IN A WM DELAY TASK delay activity Quotation withheld because this record is not cleared for redistribution. individual neurons RESPOND TO COMBINATIONS OF TASK AND STIMULUS FEATURES SUCH THAT FEATURE COMBINATIONS CANNOT BE READ OUT FROM THE RESPONSES OF THE INDIVIDUAL NEURON ALONE mixed selectivity Quotation withheld because this record is not cleared for redistribution. 3% — Proportion of delay-active lateral prefrontal neurons that fire at a stable rate through the whole delay Quotation withheld because this record is not cleared for redistribution. 40–100 Hz — Frequency range the review assigns to the gamma band whose power is sustained during working-memory delays Quotation withheld because this record is not cleared for redistribution. 42 — Number of cortical areas recorded simultaneously in the broadest-coverage monkey study of content-selective delay spiking Quotation withheld because this record is not cleared for redistribution. ~100,000 — Number of voxels in a 3D fMRI brain image, each pooling tens of thousands of neurons Quotation withheld because this record is not cleared for redistribution. 10–15 s — Typical fixed delay length used in the most common fMRI boxcar design for estimating delay activity Quotation withheld because this record is not cleared for redistribution. delay activity delay activity 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/s41583-019-0176-7.