A large-scale standardized physiological survey reveals functional organization of the mouse visual cortex

A large-scale standardized physiological survey reveals functional organization of the mouse visual cortex de Vries Lecoq Buice Groblewski Ocker Oliver Feng Cain

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 large-scale standardized physiological survey reveals functional organization of the mouse visual cortex 2020 Nature Neuroscience 10.1038/s41593-019-0550-9 de-Vries2020-a-large-scale-standardized-physiological-sur de-Vries2020-a-large-scale-standardized-physiological-sur.txt, the project's staged copy c35510c84a7d6729b95052f272c7e6984c976058 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 how stimulus representations ARE TRANSFORMED THROUGHOUT THE VISUAL CORTEX, surveyed rather than sampled Quotation withheld because this record is not cleared for redistribution. whether the standard account of visual cortex survives a survey that does not select its neurons by their responses - A KEY CHALLENGE RESULTS FROM THE FACT THAT THIS UNDERSTANDING IS BASED ON MANY SMALL STUDIES, RECORDING RESPONSES FROM DIFFERENT STAGES IN THE CIRCUIT, USING DIFFERENT STIMULI AND ANALYSES Quotation withheld because this record is not cleared for redistribution. an open survey in which THE LARGEST CLASS IS NOT RELIABLY RESPONSIVE TO ANY OF THE STIMULI AND BECOMES PROGRESSIVELY LARGER IN HIGHER VISUAL AREAS Quotation withheld because this record is not cleared for redistribution. existence Quotation withheld because this record is not cleared for redistribution. functional organization of mouse visual cortex Quotation withheld because this record is not cleared for redistribution. calcium-imaging calcium-imaging Quotation withheld because this record is not cleared for redistribution. mechanisms Hand-tuning stimuli to each recorded neuron builds selection bias into the classical literature, which is why the standard model looks better in small studies than in an unbiased survey. Quotation withheld because this record is not cleared for redistribution. mechanisms The unresponsive 'none' neurons may in fact be visually driven but tuned to highly sparse, specific natural features constructed by hierarchical processing, which the presented stimulus set never hits. Quotation withheld because this record is not cleared for redistribution. mechanisms Because responses are so sparse, different neurons are active at different moments, which is what keeps their pairwise noise correlations weak. Quotation withheld because this record is not cleared for redistribution. the bias this survey is built against, described as a procedure: NEURONS WERE IDENTIFIED BY RESPONDING TO A TEST STIMULUS - THE STIMULUS WAS THEN HAND-TUNED TO ELICIT THE STRONGEST RELIABLE RESPONSE FROM THAT NEURON, AND THE EXPERIMENT PROCEEDED USING MANIPULATIONS AROUND THIS CONDITION Quotation withheld because this record is not cleared for redistribution. and what that procedure cannot see: THESE STUDIES MAY HAVE FAILED TO CAPTURE THE VARIABILITY OF RESPONSES, THE BREADTH OF FEATURES THAT WILL ELICIT A NEURAL RESPONSE, AND THE BREADTH OF FEATURES THAT DO NOT ELICIT A RESPONSE - THIS RESULTS IN SYSTEMATIC BIAS IN THE MEASUREMENT OF NEURONS AND A CONFIRMATION BIAS REGARDING MODEL ASSUMPTIONS Quotation withheld because this record is not cleared for redistribution. the one design decision that removes it: THIS PIPELINE REDUCED CRITICAL EXPERIMENTAL BIASES BY SEPARATING QUALITY CONTROL OF DATA COLLECTION FROM RESPONSE CHARACTERIZATION Quotation withheld because this record is not cleared for redistribution. and the reason it matters for a whole literature: THIS UNDERSTANDING IS BASED ON MANY SMALL STUDIES, RECORDING RESPONSES FROM DIFFERENT STAGES IN THE CIRCUIT, USING DIFFERENT STIMULI AND ANALYSES - THE INHERENT EXPERIMENTAL SELECTION BIASES AND LACK OF STANDARDIZATION OF THIS APPROACH INTRODUCE ADDITIONAL OBSTACLES TO CREATING A COHESIVE UNDERSTANDING OF CORTICAL FUNCTION Quotation withheld because this record is not cleared for redistribution. The putative ventral higher visual areas could not be imaged, so the expected inverse (spatial-over-motion) organisation could not be tested. what this limits: the same scope as limitations[4] - the putative ventral higher visual areas could not be imaged, so the expected inverse organisation is missing for a reason about the window, not the brain Quotation withheld because this record is not cleared for redistribution. Neurons classed as responding only to natural stimuli may be conventional cells tuned to spatial or temporal frequencies the stimulus set did not sample. what this limits: the class of neurons responding only to natural stimuli: THESE NEURONS COULD BE 'CLASSICAL' NEURONS AS WELL BUT ARE TUNED FOR SPATIAL OR TEMPORAL FREQUENCIES THAT WERE NOT INCLUDED IN OUR STIMULUS SET Quotation withheld because this record is not cleared for redistribution. The wavelet models leave a large amount of visually responsive activity unexplained. what this limits: the explanatory reach of the wavelet models: THERE IS A GREAT DEAL OF VISUALLY RESPONSIVE ACTIVITY THAT IS NOT BEING CAPTURED BY THESE MODELS Quotation withheld because this record is not cleared for redistribution. The epileptiform-event screening used to exclude mice could not be applied to the inhibitory Cre lines. what this limits: the epileptiform screening, and therefore the quality control on the inhibitory lines: INHIBITORY LINES WERE EXCLUDED FROM THIS ANALYSIS AS THE NEURONAL LABELLING WAS TOO SPARSE TO RELIABLY ASSESS THESE EVENTS Quotation withheld because this record is not cleared for redistribution. one half of the expected organisation is missing for a reason that is about the window, not the brain - WE WERE UNABLE TO IMAGE THE PUTATIVE VENTRAL AREAS LI, POR, OR P WITHIN OUR CRANIAL WINDOW what this limits: reading the absence of a ventral counterpart to the dorsal specialisation as a finding about the cortex Quotation withheld because this record is not cleared for redistribution. the largest class may be an artefact of the stimulus set rather than a property of the cells - IT IS POSSIBLE THESE NEURONS ARE VISUALLY DRIVEN, BUT ARE RESPONSIVE TO HIGHLY SPARSE AND SPECIFIC NATURAL FEATURES, or THESE NEURONS COULD BE INVOLVED IN NON-VISUAL COMPUTATION what this limits: reading 'does not respond to visual stimuli' as 'is not visual' Quotation withheld because this record is not cleared for redistribution. and the stimulus parameters may have been set too low - AS OUR STIMULUS PARAMETERS WERE CHOSEN TO MATCH PREVIOUS MEASUREMENTS OF MOUSE ACUITY, THIS COULD SUGGEST THAT THE ACUITY OF MOUSE HAS BEEN UNDERESTIMATED what this limits: reading poor prediction for artificial stimuli as a property of the neurons rather than of the parameter range Quotation withheld because this record is not cleared for redistribution. the models fail for most of the cortex - EVEN WITH THE INCLUSION OF RUNNING, OUR MODELS PREDICT RESPONSES IN A MINORITY OF NEURONS what this limits: any claim that the standard models describe this cortex: EVEN WITH THE INCLUSION OF RUNNING, OUR MODELS PREDICT RESPONSES IN A MINORITY OF NEURONS Quotation withheld because this record is not cleared for redistribution. the pipeline itself - STANDARD OPERATING PROCEDURES WITH INTEGRATED ENGINEERING TOOLS, applied to make 243 mice comparable to each other Quotation withheld because this record is not cleared for redistribution. quality control of the recording separated from characterisation of the response eliminates: the dataset selecting for cells that respond - STRICT QUALITY CONTROL MEASURES NOT DEPENDENT UPON STIMULUS-DRIVEN ACTIVITY · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. the functional classes validated against a model containing the standard machinery plus running speed eliminates: the classes being an artefact of the reliability clustering - the models contain MOST OF THE BASIC FEATURES FOUND IN VISUAL NEUROPHYSIOLOGICAL MODELING as well as THE RUNNING SPEED OF THE MOUSE · outcome (stated by the tagger, no quotation): partial Quotation withheld because this record is not cleared for redistribution. both artificial and natural stimuli presented to the same neurons eliminates: a conclusion that holds only for parameterised stimuli - RESPONSES TO NATURAL SCENES CANNOT BE WELL PREDICTED BY RESPONSES TO NOISE OR GRATING STIMULI · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. running checked as an explanation for the unresponsive class eliminates: locomotion accounting for the 'none' class - WE FOUND LITTLE EVIDENCE THAT THESE NEURONS WERE CORRELATED WITH THE MOUSE'S RUNNING · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. a STANDARDISED survey - fixed operating procedures and stimuli applied identically across 59,610 neurons in 243 mice - in place of experiment-specific stimulus choices eliminates: the experimenter's choice of stimulus as the determinant of what is found. The paper states the fault it is answering: earlier studies 'may have failed to capture…', producing 'a CONFIRMATION BIAS regarding…' · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. none none Quotation withheld because this record is not cleared for redistribution. THE LARGEST CLASS OF NEURONS RESPONDS TO NOTHING: REMARKABLY, THE LARGEST CLASS OF NEURONS WAS THE “none” CLASS, CONSTITUTING THOSE NEURONS THAT DID NOT RESPOND RELIABLY TO ANY OF THE STIMULI (34% OF NEURONS) - THESE NEURONS ARE THE LEAST LIKELY TO BE DESCRIBED BY “classical receptive fields,” AS EVIDENCED BY THEIR POOR MODEL PERFORMANCE FOR ALL STIMULI Quotation withheld because this record is not cleared for redistribution. and in one area it is almost everything: AREA RL HAS THE LARGEST PROPORTION OF NEURONS IN THE “none” CLASS, OVER 85%, CONSISTENT WITH THE VERY LOW PERCENT OF RESPONSIVE NEURONS Quotation withheld because this record is not cleared for redistribution. THE PAPER ASKS THE QUESTION ITSELF AND LEAVES IT OPEN: WHAT, THEN, DO THESE NEURONS DO? - and offers both readings, that they ARE RESPONSIVE TO HIGHLY SPARSE AND SPECIFIC NATURAL FEATURES or COULD BE INVOLVED IN NON-VISUAL COMPUTATION Quotation withheld because this record is not cleared for redistribution. THE CLASS THE STANDARD MODEL EXPLAINS IS A TENTH OF THE CORTEX: NEURONS IN THE “DG-SG-NS-NM” CLASS WERE WELL PREDICTED, WITH VALUES COMPARABLE TO THOSE FOUND IN PRIMATES - HOWEVER, THIS CLASS CONSTITUTES ONLY 10% OF THE MOUSE VISUAL CORTEX. And EVEN WITH THE INCLUSION OF RUNNING, OUR MODELS PREDICT RESPONSES IN A MINORITY OF NEURONS Quotation withheld because this record is not cleared for redistribution. THE SELECTION BIAS OF A WHOLE LITERATURE, DESCRIBED AS A PROCEDURE: NEURONS WERE IDENTIFIED BY RESPONDING TO A TEST STIMULUS - THE STIMULUS WAS THEN HAND-TUNED TO ELICIT THE STRONGEST RELIABLE RESPONSE FROM THAT NEURON... THIS RESULTS IN SYSTEMATIC BIAS IN THE MEASUREMENT OF NEURONS AND A CONFIRMATION BIAS REGARDING MODEL ASSUMPTIONS Quotation withheld because this record is not cleared for redistribution. and the fix is one sentence of study design: THIS PIPELINE REDUCED CRITICAL EXPERIMENTAL BIASES BY SEPARATING QUALITY CONTROL OF DATA COLLECTION FROM RESPONSE CHARACTERIZATION - the decision to include a cell is made before anyone knows whether it responds Quotation withheld because this record is not cleared for redistribution. THE PROBLEM STATED AS A PROBLEM OF SYNTHESIS: THIS UNDERSTANDING IS BASED ON MANY SMALL STUDIES, RECORDING RESPONSES FROM DIFFERENT STAGES IN THE CIRCUIT, USING DIFFERENT STIMULI AND ANALYSES - THE INHERENT EXPERIMENTAL SELECTION BIASES AND LACK OF STANDARDIZATION OF THIS APPROACH INTRODUCE ADDITIONAL OBSTACLES TO CREATING A COHESIVE UNDERSTANDING OF CORTICAL FUNCTION Quotation withheld because this record is not cleared for redistribution. and the field asked for exactly this: INFLUENTIAL REVIEWS HAVE QUESTIONED THE VALIDITY OF THIS STANDARD MODEL, AND HAVE ARGUED THAT “What would be most helpful is to accumulate a database of single unit or multi-unit data that would allow modelers to test their best theory under ecological conditions.” Quotation withheld because this record is not cleared for redistribution. A MISSING HALF OF THE ANSWER, ATTRIBUTED TO THE WINDOW AND NOT THE BRAIN: THE LACK OF AN INVERSE RELATIONSHIP, WHEREIN SPATIAL INFORMATION IS OVERREPRESENTED RELATIVE TO MOTION IN A PUTATIVE “ventral” STREAM, LIKELY REFLECTS THE FACT THAT WE WERE UNABLE TO IMAGE THE PUTATIVE VENTRAL AREAS LI, POR, OR P WITHIN OUR CRANIAL WINDOW Quotation withheld because this record is not cleared for redistribution. TWO INTERNEURON CLASSES, 94% AGAINST 9%, TO THE SAME STIMULUS: 94% OF Sst NEURONS IN V1 ARE RESPONSIVE TO DRIFTING GRATINGS, AND RESPOND QUITE RELIABLY ACROSS TRIALS, FAR MORE THAN THE OTHER CRE LINES - Vip NEURONS, ON THE OTHER HAND, ARE LARGELY UNRESPONSIVE, EVEN SUPPRESSED, BY DRIFTING GRATINGS, WITH ONLY 9% OF Vip NEURONS IN V1 LABELLED RESPONSIVE Quotation withheld because this record is not cleared for redistribution. and the same cells are highly responsive to a different stimulus class: Vip NEURONS ARE VERY RESPONSIVE TO BOTH NATURAL SCENES AND NATURAL MOVIES, but SHOW LITTLE SELECTIVITY TO THESE STIMULI, AS THEIR MEDIAN LIFETIME SPARSENESS IS LOWER THAN BOTH THE Sst AND THE EXCITATORY NEURONS. `Vip cells are unresponsive` is true of one stimulus set and false of another Quotation withheld because this record is not cleared for redistribution. A MEASUREMENT THAT REVERSES AN EARLIER ONE, WITH THE STIMULUS NAMED: RECEPTIVE FIELD MAPPING USING LOCALLY SPARSE NOISE REVEALED THAT Vip NEURONS IN V1 HAVE REMARKABLY LARGE RECEPTIVE FIELD AREAS, LARGER THAN BOTH Sst AND EXCITATORY NEURONS, IN CONTRAST TO THE SMALLER SUMMATION AREA FOR Vip NEURONS, PREVIOUSLY MEASURED USING WINDOWED DRIFTING GRATINGS Quotation withheld because this record is not cleared for redistribution. THE STIMULUS CLASS DECIDES THE ANSWER, AND THE PAPER SAYS SO BEFORE MEASURING: IT IS KNOWN THAT STIMULUS STATISTICS AFFECT VISUAL RESPONSES, SUCH THAT RESPONSES TO NATURAL SCENES CANNOT BE WELL PREDICTED BY RESPONSES TO NOISE OR GRATING STIMULI - WHILE ARTIFICIAL STIMULI CAN BE EASILY PARAMETERIZED AND INTERPRETED, NATURAL STIMULI ARE CLOSER TO WHAT IS ETHOLOGICALLY RELEVANT Quotation withheld because this record is not cleared for redistribution. the headline positive, with its complement: 77% OF NEURONS IN THE MOUSE VISUAL CORTEX RESPOND TO AT LEAST ONE OF THESE VISUAL STIMULI - and THE RESPONSES TO ALL STIMULI ARE HIGHLY SPARSE AND VARIABLE Quotation withheld because this record is not cleared for redistribution. THE DIFFERENCES ARE BETWEEN THE INHIBITORY LINES, NOT THE EXCITATORY ONES: WHILE SUBTLE DIFFERENCES DO EXIST BETWEEN THE EXCITATORY CRE LINES, THESE POPULATIONS ARE LARGELY SIMILAR; THE MORE MARKED DIFFERENCES ARE AMONG THE INHIBITORY INTERNEURONS Quotation withheld because this record is not cleared for redistribution. and yet the excitatory lines are not interchangeable either: RORB, SCNN1A-TG3, AND NR5A1, WHICH LABEL DISTINCT LAYER 4 POPULATIONS IN V1, EXHIBIT DISTINCT SPATIAL AND TEMPORAL TUNING PROPERTIES, DIFFERENT DEGREES OF RUNNING CORRELATION, AND SUBTLE DIFFERENCES IN THEIR CLASS DISTRIBUTION - THESE DIFFERENCES SUGGEST THAT THERE ARE SEPARATE CHANNELS OF FEEDFORWARD INFORMATION Quotation withheld because this record is not cleared for redistribution. THE CLASSES DO NOT RESPECT THE AREAS: Quotation withheld Quotation withheld because this record is not cleared for redistribution. and what organisation there is comes from proportions rather than presence: THE OVERREPRESENTATION OF CLASSES RESPONSIVE TO NATURAL MOVIES AND MOTION STIMULI IN AREAS AL, PM, AND AM RELATIVE TO THE OTHER CLASSES IS CONSISTENT WITH THE ASSIGNMENT OF THESE AREAS TO THE PUTATIVE “dorsal” OR “motion” STREAM Quotation withheld because this record is not cleared for redistribution. THE DATASET IS THE CONTRIBUTION AND IT IS ALREADY BEING REUSED: THESE DATA HAVE BEEN USED BY OTHER RESEARCHERS TO DEVELOP IMAGE PROCESSING METHODS, TO EXAMINE STIMULUS ENCODING AND DECODING, AND TO TEST MODELS OF CORTICAL COMPUTATIONS 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. method: run every animal through one standardised pipeline with QUALITY CONTROL MEASURES NOT DEPENDENT UPON STIMULUS-DRIVEN ACTIVITY, so that entry into the dataset is decided before any response is measured a second reading DIFFERING from the recorded value. Recorded: calcium-imaging — 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 population of cortical neurons that respond reliably to none of the stimuli a field has spent decades building a second reading DIFFERING from the recorded value. Recorded: functional organization of mouse visual cortex — 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 → Jewell et al. L0-penalized event detection method Quotation withheld because this record is not cleared for redistribution. builds-on → Vinje and Gallant definition of lifetime sparseness Quotation withheld because this record is not cleared for redistribution. builds-on → Schoppe et al. computation of CCmax Quotation withheld because this record is not cleared for redistribution. critiques → the standard model of visual cortex built from hand-tuned single-neuron recordings differs in: how the neurons enter the sample - NEURONS WERE IDENTIFIED BY RESPONDING TO A TEST STIMULUS and THE STIMULUS WAS THEN HAND-TUNED TO ELICIT THE STRONGEST RELIABLE RESPONSE Quotation withheld because this record is not cleared for redistribution. contradicts → the smaller summation area reported for Vip neurons differs in: the stimulus used to map it - LOCALLY SPARSE NOISE here, against WINDOWED DRIFTING GRATINGS previously Quotation withheld because this record is not cleared for redistribution. extends → previous work showing functional differences among cortical areas in layer 2/3, extended across layers and 12 transgenic lines Quotation withheld because this record is not cleared for redistribution. qualifies → the assignment of AL, PM and AM to a dorsal stream differs in: that the complement was not testable - WE WERE UNABLE TO IMAGE THE PUTATIVE VENTRAL AREAS LI, POR, OR P WITHIN OUR CRANIAL WINDOW Quotation withheld because this record is not cleared for redistribution. none —is-a→ class none class strength of the relation: 1 · note: THE LARGEST CLASS OF NEURONS WAS THE 'none' CLASS, CONSTITUTING THOSE NEURONS THAT DID NOT RESPOND RELIABLY TO ANY OF THE STIMULI (34% OF NEURONS) Quotation withheld because this record is not cleared for redistribution. none —part-of→ RL none RL depends on: which area is measured - the same class is 34% of the survey as a whole · strength of the relation: 0.85 · note: AREA RL HAS THE LARGEST PROPORTION OF NEURONS IN THE 'none' CLASS, OVER 85% Audit 2026-09-28: DOUBTFUL - read against its span and kept, but RL has the largest share of non-responsive neurons; part-of is loose. Quotation withheld because this record is not cleared for redistribution. quality control —causes→ bias quality control bias strength of the relation: 1 · note: THIS PIPELINE REDUCED CRITICAL EXPERIMENTAL BIASES BY SEPARATING QUALITY CONTROL OF DATA COLLECTION FROM RESPONSE CHARACTERIZATION Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the pipeline reduced biases; the edge carries no direction. Quotation withheld because this record is not cleared for redistribution. natural —contrasted-with→ gratings natural gratings strength of the relation: 1 · note: RESPONSES TO NATURAL SCENES CANNOT BE WELL PREDICTED BY RESPONSES TO NOISE OR GRATING STIMULI Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the source endpoint 'natural' is a bare adjective; the span writes 'natural (scenes and movies) stimuli'. Quotation withheld because this record is not cleared for redistribution. correlation correlation Quotation withheld because this record is not cleared for redistribution. Kolmogorov-Smirnov test Kolmogorov-Smirnov test Quotation withheld because this record is not cleared for redistribution. Bonferroni correction Bonferroni correction Quotation withheld because this record is not cleared for redistribution. chi-squared test chi-squared test Quotation withheld because this record is not cleared for redistribution. Sidak multiple-comparisons correction Sidak multiple-comparisons correction Quotation withheld because this record is not cleared for redistribution. Gaussian mixture model clustering with Bayesian Information Criterion model selection Gaussian mixture model clustering with Bayesian Information Criterion model selection Quotation withheld because this record is not cleared for redistribution. K-nearest-neighbors classification (decoding) K-nearest-neighbors classification (decoding) Quotation withheld because this record is not cleared for redistribution. cross-validation cross-validation Quotation withheld because this record is not cleared for redistribution. ridge regression ridge regression Quotation withheld because this record is not cleared for redistribution. polynomial regression polynomial regression Quotation withheld because this record is not cleared for redistribution. coefficient of variation coefficient of variation Quotation withheld because this record is not cleared for redistribution. interquartile range interquartile range Quotation withheld because this record is not cleared for redistribution. bootstrap bootstrap Quotation withheld because this record is not cleared for redistribution. permutation/shuffle test permutation/shuffle test Quotation withheld because this record is not cleared for redistribution. lifetime sparseness lifetime sparseness Quotation withheld because this record is not cleared for redistribution. IN TOTAL, 77% OF NEURONS WERE RESPONSIVE TO AT LEAST ONE OF THE VISUAL STIMULI PRESENTED Quotation withheld because this record is not cleared for redistribution. BY FAR THE LARGEST SINGLE CLASS REVEALED BY THIS ANALYSIS IS THAT OF NEURONS THAT ARE LARGELY UNRESPONSIVE TO ALL STIMULI, TERMED `none,` WHICH CONTAINS 34±2% OF THE NEURONS Quotation withheld because this record is not cleared for redistribution. Area RL is the extreme case, with over 85% of its neurons falling into the unresponsive 'none' class. Quotation withheld because this record is not cleared for redistribution. How reliably a neuron responds to one stimulus type predicts almost nothing about how reliably it responds to another. Quotation withheld because this record is not cleared for redistribution. A standard linear-plus-quadratic Gabor wavelet encoding model fits only about 2% of the population well, with median correlations near 0.2. Quotation withheld because this record is not cleared for redistribution. Adding the mouse's running speed to the encoding model gave essentially no extra predictive power except for Vip neurons in V1. Quotation withheld because this record is not cleared for redistribution. Most neurons do not respond on most individual trials, even to the stimulus condition that drives their largest average response. Quotation withheld because this record is not cleared for redistribution. Sst interneurons are the most reliably driven population, with 94% of Sst neurons in V1 responsive to drifting gratings. Quotation withheld because this record is not cleared for redistribution. Vip interneurons are the mirror image: only 9% of Vip neurons in V1 are responsive to drifting gratings, and many are suppressed by them. Quotation withheld because this record is not cleared for redistribution. COMPARISONS ACROSS AREAS AND LAYERS REVEALED THAT DIRECTION SELECTIVITY IS HIGHEST IN LAYER 4 OF V1 Quotation withheld because this record is not cleared for redistribution. Excitatory neurons had a median lifetime sparseness of 0.77 for natural scenes. Quotation withheld because this record is not cleared for redistribution. THIS DIVERGENCE IN REPRESENTATION OF THE MOTION CLASSES FROM THE NATURAL CLASSES IN AREAS AL, PM, AND AM IS CONSISTENT WITH THE PUTATIVE DORSAL AND VENTRAL STREAM SEGREGATION IN THE VISUAL CORTEX Quotation withheld because this record is not cleared for redistribution. Vip neurons in V1 have unusually large receptive fields mapped with sparse noise, larger than Sst or excitatory neurons and at odds with their small summation area measured with windowed gratings. Quotation withheld because this record is not cleared for redistribution. REMARKABLY, THE LARGEST CLASS OF NEURONS WAS THE `none` CLASS, CONSTITUTING THOSE NEURONS THAT DID NOT RESPOND RELIABLY TO ANY OF THE STIMULI (34% OF NEURONS) Quotation withheld because this record is not cleared for redistribution. AREA RL HAS THE LARGEST PROPORTION OF NEURONS IN THE `none` CLASS, OVER 85% Quotation withheld because this record is not cleared for redistribution. the well-predicted class CONSTITUTES ONLY 10% OF THE MOUSE VISUAL CORTEX, and EVEN WITH THE INCLUSION OF RUNNING, OUR MODELS PREDICT RESPONSES IN A MINORITY OF NEURONS Quotation withheld because this record is not cleared for redistribution. 77% OF NEURONS IN THE MOUSE VISUAL CORTEX RESPOND TO AT LEAST ONE OF THESE VISUAL STIMULI, and THE RESPONSES TO ALL STIMULI ARE HIGHLY SPARSE AND VARIABLE Quotation withheld because this record is not cleared for redistribution. 94% OF Sst NEURONS IN V1 ARE RESPONSIVE TO DRIFTING GRATINGS, while ONLY 9% OF Vip NEURONS IN V1 are LABELLED RESPONSIVE Quotation withheld because this record is not cleared for redistribution. Vip NEURONS IN V1 HAVE REMARKABLY LARGE RECEPTIVE FIELD AREAS, LARGER THAN BOTH Sst AND EXCITATORY NEURONS, IN CONTRAST TO THE SMALLER SUMMATION AREA PREVIOUSLY MEASURED USING WINDOWED DRIFTING GRATINGS Quotation withheld because this record is not cleared for redistribution. Quotation withheld Quotation withheld because this record is not cleared for redistribution. WHILE SUBTLE DIFFERENCES DO EXIST BETWEEN THE EXCITATORY CRE LINES, THESE POPULATIONS ARE LARGELY SIMILAR; THE MORE MARKED DIFFERENCES ARE AMONG THE INHIBITORY INTERNEURONS Quotation withheld because this record is not cleared for redistribution. lifetime sparseness lifetime sparseness Quotation withheld because this record is not cleared for redistribution. population sparseness population sparseness Quotation withheld because this record is not cleared for redistribution. reliability reliability Quotation withheld because this record is not cleared for redistribution. response class response class Quotation withheld because this record is not cleared for redistribution. imaging depths imaging depths Quotation withheld because this record is not cleared for redistribution. mean response mean response Quotation withheld because this record is not cleared for redistribution. tuning metrics tuning metrics Quotation withheld because this record is not cleared for redistribution. running state running state Quotation withheld because this record is not cleared for redistribution. pupil position and area pupil position and area Quotation withheld because this record is not cleared for redistribution. running running Quotation withheld because this record is not cleared for redistribution. class class Quotation withheld because this record is not cleared for redistribution. 59,610 — neurons imaged in the survey, drawn from 432 experiments Quotation withheld because this record is not cleared for redistribution. 77% — share of neurons responsive to at least one visual stimulus Quotation withheld because this record is not cleared for redistribution. 34±2% — share of clustered neurons in the 'none' class, unresponsive to every stimulus Quotation withheld because this record is not cleared for redistribution. 33% — share of neurons in area RL responding to any stimulus, the lowest of the six areas Quotation withheld because this record is not cleared for redistribution. median 0.36 — WHILE Sst NEURONS WERE COMPARABLE TO EXCITATORY NEURONS (MEDIAN 0.77), Vip NEURONS EXHIBITED LOW LIFETIME SPARSENESS (MEDIAN 0.36) Quotation withheld because this record is not cleared for redistribution. 13% — share of neurons whose evoked responses were significantly modulated by running Quotation withheld because this record is not cleared for redistribution. 1.9 fold — increase in the median evoked response when the mouse was running versus stationary Quotation withheld because this record is not cleared for redistribution. median 0.02 — noise correlation of excitatory populations, against 0.06 for Sst and 0.15 for Vip Quotation withheld because this record is not cleared for redistribution. 517451 — number of free parameters in the 3D Gabor wavelet encoding model Quotation withheld because this record is not cleared for redistribution. 33.6% — share of neurons successfully matched across all three imaging sessions Quotation withheld because this record is not cleared for redistribution. 21% — share of responsive neurons whose trial responses the average tuning curve predicts with r above 0.5 (11% of all neurons) Quotation withheld because this record is not cleared for redistribution. over 85% — share of area RL neurons in the unresponsive 'none' class Quotation withheld because this record is not cleared for redistribution. 34% — of neurons in the 'none' class - the largest single class in the survey depends on: the area - AREA RL HAS THE LARGEST PROPORTION OF NEURONS IN THE 'none' CLASS, OVER 85% Quotation withheld because this record is not cleared for redistribution. only 10% — of the mouse visual cortex in the class the standard model predicts well Quotation withheld because this record is not cleared for redistribution. 77% — of neurons responding to AT LEAST ONE of the stimuli - so 23% respond to none of them Quotation withheld because this record is not cleared for redistribution. 94% — of Sst neurons in V1 responsive to drifting gratings depends on: the cell class - the matched Vip figure is 9% Quotation withheld because this record is not cleared for redistribution. over 85% — of neurons in area RL falling in the 'none' class 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/s41593-019-0550-9.