A Neural Correlate of Working Memory in the Monkey Primary Visual Cortex

A Neural Correlate of Working Memory in the Monkey Primary Visual Cortex Supèr Spekreijse Lamme

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.

A Neural Correlate of Working Memory in the Monkey Primary Visual Cortex 2001 Science 10.1126/science.1060496 sup-r2001-a-neural-correlate-of-working-memo sup-r2001-a-neural-correlate-of-working-memo.txt, the project's staged copy ddaae539d4b79d17569697987115fe65b94e5d61 open-access visual-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 existence Quotation withheld because this record is not cleared for redistribution. visual working memory in primary visual cortex Quotation withheld because this record is not cleared for redistribution. electrophysiology electrophysiology Quotation withheld because this record is not cleared for redistribution. mechanisms Figure-ground contextual modulation in V1, generated by recurrent feedback, is carried on past stimulus offset and thereby converts the sensory representation into a working-memory trace. Quotation withheld because this record is not cleared for redistribution. mechanisms Two processes act on V1 during the delay in opposite directions: one suppresses overall firing while a second preserves the figure-versus-ground difference. Quotation withheld because this record is not cleared for redistribution. The signal recorded in V1 is stated to depend on feedback from higher areas, so the site of the recording is not the site of the process. what this limits: this result as a finding about primary visual cortex: the modulation depends on feedback from higher visual areas, which the authors say makes it a specific correlate of recurrent processing rather than of local V1 computation Quotation withheld because this record is not cleared for redistribution. The early figure-ground signal is the same whether the animal later remembers or forgets; only the late persistence separates them. what this limits: an early V1 figure-ground measurement as an index of what will be remembered: segregation is equally good on correct and incorrect trials, and the two only diverge later in the delay Quotation withheld because this record is not cleared for redistribution. The memory signal decays with the length of the delay, so its strength depends on how long the animal is asked to wait. what this limits: a modulation strength quoted from this paper: it is related to the delay period and becomes weaker for longer delays, so the figure is only comparable at a matched delay Quotation withheld because this record is not cleared for redistribution. the whole delayed-response experiment repeated with a STATIC orientation-defined figure in place of the moving one, followed by a mask eliminates: any peculiarity of the MOVING stimulus as the source of the sustained contextual modulation - motion has its own persistence, so a modulation that outlasts a moving figure could be an aftereffect of the motion rather than a memory trace · outcome: held Quotation withheld because this record is not cleared for redistribution. contextual modulation contextual modulation Quotation withheld because this record is not cleared for redistribution. recurrent processing recurrent processing Quotation withheld because this record is not cleared for redistribution. working memory working memory Quotation withheld because this record is not cleared for redistribution. delayed-response task delayed-response task Quotation withheld because this record is not cleared for redistribution. receptive fields receptive fields Quotation withheld because this record is not cleared for redistribution. contextual modulation - AS A CORRELATE OF RECURRENT PROCESSING: a V1 signal whose source is feedback from higher visual areas, so it indexes a loop rather than the area it is recorded in Quotation withheld because this record is not cleared for redistribution. the memory signal - DEFINED BY THE ANIMAL'S REPORT: trials are sorted after the fact into correct and incorrect by whether the saccade landed on the remembered location within 500 ms, and the neural difference is computed between those two sets Quotation withheld because this record is not cleared for redistribution. the figure-ground difference - NOT STIMULUS-DRIVEN BY CONSTRUCTION: the motion on the receptive field was arranged to be identical on average for figure and ground, so the difference cannot be attributed to the local stimulus Quotation withheld because this record is not cleared for redistribution. explains-away → fMRI studies finding little delay activation in early visual areas (ref. 15) differs in: WHETHER THE MEASURE CAN SEE THE SIGNAL. Two mechanisms run in opposite directions: 'one mechanism suppresses overall activity in V1 during the delay, Quotation withheld.' A method reading overall activity therefore records a decrease while the memory signal - a DIFFERENCE between two conditions - is maintained. The paper offers this as why 'very little delay activation is found in early visual areas' in imaging Quotation withheld because this record is not cleared for redistribution. distinguishes-from → delay-period activity in temporal (ref. 13) and prefrontal (ref. 14) cortex differs in: WHAT IS MAINTAINED DURING THE DELAY. Elsewhere it is the firing rate; here the rate falls and what persists is a CONTRAST between figure and ground - 'In that sense, Quotation withheld OR PREFRONTAL CORTEX.' Two things called delay activity that a rate measure would score in opposite directions Quotation withheld because this record is not cleared for redistribution. builds-on → the criterion that stored information is held only if needed (ref. 11) differs in: WHETHER PERSISTENCE ALONE DEMONSTRATES STORAGE. The distractor condition supplies the missing half: modulation persists when the figure is the target and decays when the same figure is the distractor, so 'the continuation of modulation is Quotation withheld of information needed for the animal's goal'. The same stimulus with the same retinal input yields opposite signals according to task relevance Quotation withheld because this record is not cleared for redistribution. recurrent processing —causes→ contextual modulation recurrent processing contextual modulation strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span says contextual modulation depends on feedback, which IMPLIES it correlates with recurrent processing; causes is loose. Quotation withheld because this record is not cleared for redistribution. working memory —operationalised-as→ delayed-response task working memory delayed-response task strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span describes the delayed-response task used for the working memory of the previous sentence; operationalised-as is loose. Quotation withheld because this record is not cleared for redistribution. p-value p-value Quotation withheld because this record is not cleared for redistribution. regression regression Quotation withheld because this record is not cleared for redistribution. standard error of the mean standard error of the mean about this quote: DAMAGED GLYPHS IN THIS QUOTE: (cid:222). These are raw PDF font glyph indices that leaked out in place of the characters they stand for - usually a mathematical operator. The quote is verbatim to the staged text and its anchor is sound; the STAGED TEXT is what is damaged. NOT repaired: the cid-to-character mapping is defined by the font embedded in that particular PDF, so guessing it would produce a clean and unverifiable quote. Repair needs the source PDF and is a re-staging job. Quotation withheld because this record is not cleared for redistribution. mean mean about this quote: DAMAGED GLYPHS IN THIS QUOTE: (cid:222). These are raw PDF font glyph indices that leaked out in place of the characters they stand for - usually a mathematical operator. The quote is verbatim to the staged text and its anchor is sound; the STAGED TEXT is what is damaged. NOT repaired: the cid-to-character mapping is defined by the font embedded in that particular PDF, so guessing it would produce a clean and unverifiable quote. Repair needs the source PDF and is a re-staging job. Quotation withheld because this record is not cleared for redistribution. After the figure was removed from the screen, V1 responses to the figure stayed stronger than responses to the ground throughout the delay, so figure-ground contextual modulation continues in the absence of the stimulus. Quotation withheld because this record is not cleared for redistribution. The strength of the maintained figure-ground modulation shrinks as the required memory delay gets longer. Quotation withheld because this record is not cleared for redistribution. On correct and incorrect trials alike, the figure was equally well segregated from the ground in the first 250 ms. Quotation withheld because this record is not cleared for redistribution. In the late part of the delay, contextual modulation vanished entirely on trials where the monkey later made the wrong saccade, while it continued on trials answered correctly. Quotation withheld because this record is not cleared for redistribution. The continuation of modulation is not passive but reflects active storage of information the animal needs for its goal. Quotation withheld because this record is not cleared for redistribution. During the delay V1 shows two dissociable effects at once: overall firing is suppressed while the figure-versus-ground difference is preserved. Quotation withheld because this record is not cleared for redistribution. The maintained activity is highly unlikely to be a neural correlate of visual persistence. Quotation withheld because this record is not cleared for redistribution. The monkeys detected the figure well overall but their performance dropped as the delay lengthened, confirming the task taxes short-term memory. Quotation withheld because this record is not cleared for redistribution. The same phenomenon appeared in a control experiment using a static rather than a moving stimulus. Quotation withheld because this record is not cleared for redistribution. contextual modulation contextual modulation Quotation withheld because this record is not cleared for redistribution. visual persistence visual persistence Quotation withheld because this record is not cleared for redistribution. relative degree of modulation persistence relative degree of modulation persistence Quotation withheld because this record is not cleared for redistribution. correct trial correct trial Quotation withheld because this record is not cleared for redistribution. latency of the cue time latency of the cue time Quotation withheld because this record is not cleared for redistribution. strength of contextual modulation strength of contextual modulation Quotation withheld because this record is not cleared for redistribution. directions of motion directions of motion Quotation withheld because this record is not cleared for redistribution. 28 ms — how long the motion-defined figure was shown before it had to be remembered Quotation withheld because this record is not cleared for redistribution. 0 and 2000 ms — range over which the memory delay (cue-time latency) was varied Quotation withheld because this record is not cleared for redistribution. P , 1027 — significance of the decline in modulation strength across the delay for the 1000 and 2000 ms delays Quotation withheld because this record is not cleared for redistribution. P , 0.0005 — significance of the difference in remaining modulation between figure-as-target and figure-as-distractor Quotation withheld because this record is not cleared for redistribution. 900 ms — length of the memory period over which modulation persisted in the static-texture control experiment Quotation withheld because this record is not cleared for redistribution. P , 1025 — significance of the correct-versus-incorrect difference in modulation in the final 250 ms before cue time (initial 250 ms did not differ, P . 0.1) Quotation withheld because this record is not cleared for redistribution. two monkeys — number of animals from which V1 multiunit activity was recorded 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/10.1126/science.1060496.