Thirst regulates motivated behavior through modulation of brainwide neural population dynamics

Thirst regulates motivated behavior through modulation of brainwide neural population dynamics | Science Allen Chen Pichamoorthy Tien Pachitario Luo Deisseroth

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.

Thirst regulates motivated behavior through modulation of brainwide neural population dynamics | Science 2019 10.1126/science.aav3932 allen2019-thirst-regulates-motivated-behavio allen2019-thirst-regulates-motivated-behavio.txt, the project's staged copy fbf8f4aea5db6ee437db5e90d127fc9c824f6dbd open-access 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 a physiological need state is represented across the whole brain and how that representation changes what a sensory cue does Quotation withheld because this record is not cleared for redistribution. how thirst motivation is represented downstream of the dedicated thirst neurons, and how that state regulates sensation and action Quotation withheld because this record is not cleared for redistribution. a brainwide mode of population activity that encodes motivational state, exists before the cue, and gates whether sensory input propagates into behaviour Quotation withheld because this record is not cleared for redistribution. mechanism Quotation withheld because this record is not cleared for redistribution. thirst motivational state Quotation withheld because this record is not cleared for redistribution. electrophysiology electrophysiology Quotation withheld because this record is not cleared for redistribution. mechanisms The thirst state sets a brainwide landscape of initial conditions, so the same sensory perturbation launches a different activity trajectory depending on the state the brain starts in Quotation withheld because this record is not cleared for redistribution. mechanisms Thirst imposes persistent pre-cue activity that modulates local circuitry in parallel across the brain, gating whether sensory input is transformed into motor output Quotation withheld because this record is not cleared for redistribution. mechanisms Hypothalamic thirst-neuron activity is broadcast to downstream circuits through successive fast synaptic transmission and/or neuromodulator release Quotation withheld because this record is not cleared for redistribution. the thirst neuron activity is broadcast to downstream circuits, possibly through multiple levels of fast synaptic transmission or via neuromodulators Quotation withheld because this record is not cleared for redistribution. sensory regions keep responding to the cue whatever the state; what satiety removes is the propagation of that signal past them to motor circuits Quotation withheld because this record is not cleared for redistribution. the state is described as a landscape of initial conditions spread across many regions that determines the trajectory a stimulus sets off Quotation withheld because this record is not cleared for redistribution. in the sated animal an energetic barrier stops the same stimulus from driving the dynamics down the path to behaviour Quotation withheld because this record is not cleared for redistribution. satiety does not turn down all cue-responsive neurons together; it acts on particular subclasses of them Quotation withheld because this record is not cleared for redistribution. The functional relevance of the widespread activity patterns is not established what this limits: the brainwide activity result: ALTHOUGH THIS ACTIVITY WAS EXTREMELY WIDESPREAD, THE FUNCTIONAL RELEVANCE OF THESE GLOBAL ACTIVITY PATTERNS REMAINS TO BE DETERMINED Quotation withheld because this record is not cleared for redistribution. Probably only a subset of the responsive regions is actually necessary for the task what this limits: any necessity claim from the imaging: IT IS LIKELY THAT ONLY A SUBSET OF BRAIN REGIONS ARE NECESSARY FOR TASK PERF[ormance] Quotation withheld because this record is not cleared for redistribution. Whether the identified mode is thirst-specific or a general arousal signal is unresolved what this limits: the thirst-specificity of the mode: IT REMAINS TO BE SEEN WHETHER ... THE THIRST-RELATED MODE WE OBSERVE REPRESENTS A MORE GENERAL FORM [of arousal] Quotation withheld because this record is not cleared for redistribution. the functional relevance of the brainwide activity is left undetermined even though its existence is the headline what this limits: the same scope as limitations[0] - the existence is the headline, the relevance is open Quotation withheld because this record is not cleared for redistribution. the authors expect only a subset of brain regions to be necessary for the task, so widespread activity is not widespread necessity what this limits: which regions are NECESSARY, as against where activity is found - a narrower scope than item 3: the authors expect only a subset of brain regions to be required for the task, so brainwide activity is not brainwide necessity Quotation withheld because this record is not cleared for redistribution. the Cue and Behavior modes are named for one thing but may carry reward anticipation, decision-making and motor preparation as well what this limits: the names on the two modes, which are hypotheses about what loaded onto them: THE CUE MODE WAS DESIGNED TO CAPTURE ACTIVITY RELATED TO THE SENSORY RESPONSE BUT COULD ALSO INCLUDE SIGNALS RELATED TO REWARD ANTICIPATION, DECISION-MAKING, AND MOTOR PREPARATION Quotation withheld because this record is not cleared for redistribution. whether other need states use different modes, or whether this one is general arousal, is left open duplicate of: limitations[2] · what this limits: the same scope as limitations[2] - whether other need states use different modes, or whether this one is general arousal, is left open Quotation withheld because this record is not cleared for redistribution. stimulation did not restore the thirsty reading everywhere: medial prefrontal and retrosplenial cortex went on decoding sated what this limits: the completeness of the state restoration: SEVERAL OF THE HIGHEST-LEVEL CORTICAL REGIONS [medial prefrontal and retrosplenial] CONTINUED TO DECODE THAT THE ANIMAL WAS SATED Quotation withheld because this record is not cleared for redistribution. initial conditions, trajectories and an energetic barrier used as the description of a motivational state Quotation withheld because this record is not cleared for redistribution. the low dimensionality and pervasiveness of brainwide cellular activity during behavior, established in Caenorhabditis elegans and larval zebrafish Quotation withheld because this record is not cleared for redistribution. spike sorting run through a framework with EXPLICIT DRIFT TRACKING AND CORRECTION eliminates: electrode drift across the session being read as a change in a neuron's firing - the probe moves relative to the tissue, so the same unit sorted early and late may not be the same unit · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. the State mode is not a discovered variable but an axis constructed to maximally separate thirsty from sated during the baseline epoch, and the Cue and Behavior modes are defined the same way from their own epochs Quotation withheld because this record is not cleared for redistribution. the mode names are declared impure by the authors themselves: the Cue mode may carry decision and motor-preparation signals, so the label names the epoch it was defined in, not the content it holds Quotation withheld because this record is not cleared for redistribution. near-orthogonality of the three axes is MEASURED before it is imposed - 82, 84 and 65 degrees prior to orthogonalisation Quotation withheld because this record is not cleared for redistribution. a single unit here is an identity maintained across an hours-long session by drift-tracking software, not by a stable observation Quotation withheld because this record is not cleared for redistribution. the brain-region label on every neuron is derived from the maximal channel on the probe and a registration to a common atlas, so region membership is an inference rather than an observation Quotation withheld because this record is not cleared for redistribution. gating here means diminished propagation beyond sensory regions, not diminished sensory response - the sensory regions themselves keep their fidelity Quotation withheld because this record is not cleared for redistribution. method: high-density extracellular recording across many brain regions during a Go/No-Go task, with population-mode decomposition, linear decoding and optogenetic stimulation a second reading DIFFERING from the recorded value. Recorded: electrophysiology — 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: state-dependent gating of a water-predicting cue: the same odour produces a brainwide wave when thirsty and only a transient when sated a second reading DIFFERING from the recorded value. Recorded: thirst motivational state — 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 → prior brainwide cellular-activity studies in Caenorhabditis elegans (ref 39) and larval zebrafish (ref 40) Quotation withheld because this record is not cleared for redistribution. builds-on → a previous large-scale mouse neocortex study using the same task (ref 23) Quotation withheld because this record is not cleared for redistribution. contradicts → the account in which satiety modulates cue responses as a broad change in responsiveness across all cue-responsive neurons Quotation withheld because this record is not cleared for redistribution. extends → recordings in hungry animals showing gating of responses to food-predicting stimuli in a variety of brain regions Quotation withheld because this record is not cleared for redistribution. supports → the classical drive-reduction model of water-seeking behaviour Quotation withheld because this record is not cleared for redistribution. builds-on → whole-brain cellular recordings in C. elegans and larval zebrafish showing low-dimensional pervasive dynamics Quotation withheld because this record is not cleared for redistribution. distinguishes-from → the view that a widespread activity pattern is thereby a necessary one status: the paper's own words: remains to be determined · differs in: THE AXIS IS WIDESPREAD AGAINST NECESSARY. 'Quotation withheld As was shown in a previous study ..., it is likely that only a subset of brain regions are necessary for task performance.' The paper's own headline observation is separated from the inference a reader would draw from it, BY THE PAPER, before anyone else can draw it. Extent is measured; necessity is not, and the two are different experiments. Quotation withheld because this record is not cleared for redistribution. satiety state —operationalised-as→ State mode satiety state State mode strength of the relation: 1 · note: the state is read off one axis of population activity that persists before odour onset Quotation withheld because this record is not cleared for redistribution. State mode —contrasted-with→ Cue mode State mode Cue mode strength of the relation: 1 · note: CRITERION: the State mode stays elevated through both trial types while the Cue mode discriminates them and collapses on satiety Quotation withheld because this record is not cleared for redistribution. Cue mode —distinct-from→ Behavior mode Cue mode Behavior mode strength of the relation: 0.7 · note: the authors declare the separation imperfect: both may carry reward anticipation, so the names are epoch labels, not content labels Quotation withheld because this record is not cleared for redistribution. odor epoch —operationalised-as→ Cue mode odor epoch Cue mode strength of the relation: 1 · note: each mode is defined by the epoch it maximally separates in - the definition is temporal, not anatomical Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the Cue mode is DEFINED over the odor epoch; operationalised-as runs from the epoch and is loose. Quotation withheld because this record is not cleared for redistribution. baseline epoch —operationalised-as→ State mode baseline epoch State mode strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the State mode is DEFINED over the baseline epoch; operationalised-as runs from the epoch and is loose. Quotation withheld because this record is not cleared for redistribution. gating —causes→ motor circuits gating motor circuits strength of the relation: 1 · note: SIGN: negative - the gating prevents the activation of motor circuits Audit 2026-09-28: DOUBTFUL - read against its span and kept, but satiety gating diminishes propagation to motor circuits; causes is loose and unsigned. Quotation withheld because this record is not cleared for redistribution. PIR —part-of→ sensory regions PIR sensory regions strength of the relation: 1 · note: and correspondingly PIR keeps decoding odour identity in all four blocks Quotation withheld because this record is not cleared for redistribution. Nos1+ —part-of→ SFO Nos1+ SFO strength of the relation: 1 · note: the causal handle is a genetically specified subset, not a region Quotation withheld because this record is not cleared for redistribution. single units —part-of→ brain region single units brain region strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but single units were recorded FROM many brain regions; part-of is loose. Quotation withheld because this record is not cleared for redistribution. thirst —operationalised-as→ Go/No-Go thirst Go/No-Go strength of the relation: 1 · note: thirst is measured by whether the mouse still works for water in this task Audit 2026-09-28: DOUBTFUL - read against its span and kept, but a thirst-motivated Go/No-Go task was used; operationalised-as is loose. Quotation withheld because this record is not cleared for redistribution. Wilcoxon signed-rank test Wilcoxon signed-rank test Quotation withheld because this record is not cleared for redistribution. Wilcoxon rank-sum test Wilcoxon rank-sum test 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. t-test with false-discovery-rate correction t-test with false-discovery-rate correction Quotation withheld because this record is not cleared for redistribution. correlation against shuffled data with Bonferroni correction correlation against shuffled data with Bonferroni correction Quotation withheld because this record is not cleared for redistribution. linear decoder (classification analysis) linear decoder (classification analysis) Quotation withheld because this record is not cleared for redistribution. unsupervised clustering unsupervised clustering Quotation withheld because this record is not cleared for redistribution. held-out test data (out-of-sample validation) held-out test data (out-of-sample validation) Quotation withheld because this record is not cleared for redistribution. standard error of the mean standard error of the mean Quotation withheld because this record is not cleared for redistribution. Wilcoxon signed-rank test Wilcoxon signed-rank test Quotation withheld because this record is not cleared for redistribution. two-tailed t test two-tailed t test Quotation withheld because this record is not cleared for redistribution. false discovery rate (FDR) correction false discovery rate (FDR) correction Quotation withheld because this record is not cleared for redistribution. EVIDENCE CORRECTED (RE-GROUNDED - THE POINTER WAS WRONG, NOT THE CLAIM) · About 24,000 neurons were recorded across 34 brain regions during thirst-motivated choice, over 87 sessions from 21 mice. EVIDENCE CORRECTED: RE-GROUNDED - THE POINTER WAS WRONG, NOT THE CLAIM. The old evidence was verbatim, anchored and correctly cut, and it contained NONE of this claim's numbers (000, 24, 34). The paper states the claim in one sentence elsewhere and the tagger did not use it. the project, 2026-08-23: 'WE HAVE SPENT THE DAY VERIFYING THAT QUOTES ARE REAL AND NONE OF IT ASKS WHETHER THEY ARE RELEVANT.' Confirmed by reading the claim, the old span and the candidate together; the candidate carries every number AND the claim's own content words. Quotation withheld because this record is not cleared for redistribution. The rewarded odour cue triggered a wave of activity that started in olfactory regions about 150 ms after odour onset and reached every recorded brain region within about 300 ms. Quotation withheld because this record is not cleared for redistribution. Satiety gated the wave: once mice were sated the identical odour cue produced only a brief transient change in firing and no licking. Quotation withheld because this record is not cleared for redistribution. Spontaneous pre-stimulus baseline firing, not just cue-evoked firing, shifted with satiety state in neurons across many brain regions. Quotation withheld because this record is not cleared for redistribution. In sated mice, odour decoding fell to near chance across much of the brain. Quotation withheld because this record is not cleared for redistribution. Single neurons were specialised: each mostly encoded one task aspect (state, cue, or behaviour), while every brain region held a mixture of all these neuron types. Quotation withheld because this record is not cleared for redistribution. Optogenetic stimulation of hypothalamic thirst neurons in sated mice pushed behaviour, within-trial brainwide dynamics, and single-neuron satiety correlates back to their thirsty-state values. Quotation withheld because this record is not cleared for redistribution. Re-creating the thirst representation in only a subset of brain regions was enough to restore thirst-motivated sensory processing, behaviour and the associated brainwide dynamics. Quotation withheld because this record is not cleared for redistribution. the Go cue produced a wave of activity that began in olfactory regions within about 150 ms and reached every recorded region within about 300 ms Quotation withheld because this record is not cleared for redistribution. the initial onset and the brainwide propagation of the activity wave both preceded the onset of licking; first licks were measured at 0.86 s after odor onset Quotation withheld because this record is not cleared for redistribution. a large fraction of units in every brain region, 54% of total or 12,986, were modulated by the task Quotation withheld because this record is not cleared for redistribution. in sated mice the cues still produced a diminished transient response but no persistent activity after the cue period Quotation withheld because this record is not cleared for redistribution. 35 percent of neurons had spontaneous baseline firing correlated with satiety state, in every brain region Quotation withheld because this record is not cleared for redistribution. satiety modulation was not a broad change across all cue-responsive neurons but acted on particular subclasses Quotation withheld because this record is not cleared for redistribution. state-related clusters were barely modulated by the cue and were active throughout the trial on thirsty or on sated trials Quotation withheld because this record is not cleared for redistribution. most neurons in most regions were selective for only one of state, cue or behaviour Quotation withheld because this record is not cleared for redistribution. the three modes together captured about half the trial-averaged task-related variance across the brain Quotation withheld because this record is not cleared for redistribution. ASSUMPTION MEASURED BEFORE IT WAS IMPOSED: the three axes were already near-orthogonal before orthogonalisation, at 82, 84 and 65 degrees Quotation withheld because this record is not cleared for redistribution. the State mode stayed elevated through Go and No-Go trials alike while the animal was thirsty Quotation withheld because this record is not cleared for redistribution. ROBUSTNESS: the same mode dynamics appeared with only 20 neurons subsampled per region and in each animal separately Quotation withheld because this record is not cleared for redistribution. on rare thirsty Go trials without a reward the Cue and State dynamics were normal and only the Behavior mode was reduced Quotation withheld because this record is not cleared for redistribution. in sated mice odour decoding fell to near chance across much of the brain, significantly so in 18 of 29 regions Quotation withheld because this record is not cleared for redistribution. optogenetic stimulation of the thirst neurons restored the behaviour of natural thirst, and ending it returned the animals to sated Quotation withheld because this record is not cleared for redistribution. REVERSIBILITY BUILT INTO THE DESIGN: a washout block with the stimulation switched off follows every stimulation block Quotation withheld because this record is not cleared for redistribution. CONTROL NULL: the Hit rate was not significantly different between Thirsty and Stim conditions, 97 versus 94 percent, P = 0.67 Quotation withheld because this record is not cleared for redistribution. stimulation globally restored the within-trial firing patterns of task-modulated neurons, and stopping it immediately returned them to sated dynamics Quotation withheld because this record is not cleared for redistribution. about two thirds of state-modulated neurons reversed their satiety-related change in baseline firing under stimulation Quotation withheld because this record is not cleared for redistribution. stimulation moved global activity specifically along the naturally defined State mode Quotation withheld because this record is not cleared for redistribution. the restored State mode level matched the natural level near the end of the thirsty block rather than its maximum Quotation withheld because this record is not cleared for redistribution. AMBIGUOUS AGGREGATE, STATED AS A FORK: a State mode matching the natural level toward the end of the thirsty trials was said to mean either a submaximal thirst state throughout the brain or a full thirst state in only a subset of regions Quotation withheld because this record is not cleared for redistribution. THE FORK IS RESOLVED BY DISAGGREGATION: per-region decoders trained on natural thirsty and sated data were applied to the stimulation trials Quotation withheld because this record is not cleared for redistribution. orbitofrontal, agranular insular and secondary motor cortex decoded the stimulated animal as very likely thirsty Quotation withheld because this record is not cleared for redistribution. medial prefrontal and retrosplenial cortex went on decoding sated, which the authors attribute to contextual or working-memory access to recent drinking Quotation withheld because this record is not cleared for redistribution. the answer to the fork: recapitulating thirst in a subset of regions is sufficient to restore the behaviour and the brainwide dynamics Quotation withheld because this record is not cleared for redistribution. piriform and olfactory cortex kept decoding odour identity accurately in all four blocks even as their state decoding shifted Quotation withheld because this record is not cleared for redistribution. 23,881 single units were recorded from 34 regions across 87 sessions in 21 mice Quotation withheld because this record is not cleared for redistribution. DESIGN CONSTRAINT: one probe per session and each region recorded at most once per mouse, so tracks map unambiguously to sessions Quotation withheld because this record is not cleared for redistribution. motivational drive motivational drive Quotation withheld because this record is not cleared for redistribution. State mode State mode Quotation withheld because this record is not cleared for redistribution. drive reduction drive reduction Quotation withheld because this record is not cleared for redistribution. initial conditions initial conditions Quotation withheld because this record is not cleared for redistribution. Go/No-Go Go/No-Go Quotation withheld because this record is not cleared for redistribution. Cue mode Cue mode Quotation withheld because this record is not cleared for redistribution. Behavior mode Behavior mode Quotation withheld because this record is not cleared for redistribution. optogenetic stimulation optogenetic stimulation Quotation withheld because this record is not cleared for redistribution. single units single units Quotation withheld because this record is not cleared for redistribution. licking behavior licking behavior Quotation withheld because this record is not cleared for redistribution. satiety state satiety state Quotation withheld because this record is not cleared for redistribution. State mode State mode Quotation withheld because this record is not cleared for redistribution. decoding decoding Quotation withheld because this record is not cleared for redistribution. ~24,000 — neurons whose activity was recorded, spanning 34 brain regions, 87 sessions and 21 mice Quotation withheld because this record is not cleared for redistribution. 23,881 — exact number of single units isolated, about 255 recorded simultaneously per session Quotation withheld because this record is not cleared for redistribution. 54% of total; 12,986 — proportion and count of recorded units whose firing was modulated by the task Quotation withheld because this record is not cleared for redistribution. N = 8395,35% of total — count and proportion of neurons whose baseline firing correlated with satiety state Quotation withheld because this record is not cleared for redistribution. ~50% — share of trial-averaged task-related variance captured by the three orthogonal activity modes together Quotation withheld because this record is not cleared for redistribution. 82° and 84° — angles between the State axis and the Cue and Behavior axes before orthogonalisation (65° between Cue and Behavior) Quotation withheld because this record is not cleared for redistribution. 1195 of 1781,67% — share of neurons that lowered their baseline rate on satiation and then raised it again under optogenetic thirst stimulation Quotation withheld because this record is not cleared for redistribution. 97 ± 1% versus 94 ± 2% — hit rate during natural thirst versus during optogenetic stimulation of thirst neurons, statistically indistinguishable Quotation withheld because this record is not cleared for redistribution. ~24,000 neurons in 34 brain regions — scale of the recording behind every brainwide claim Quotation withheld because this record is not cleared for redistribution. 54% of total; 12,986 — share of recorded units modulated by the task Quotation withheld because this record is not cleared for redistribution. N = 8395,35% of total — share of units whose spontaneous baseline firing tracked satiety state Quotation withheld because this record is not cleared for redistribution. ~150 ms to ~300 ms — onset in olfactory regions and spread to every recorded region Quotation withheld because this record is not cleared for redistribution. 0.86 ± 0.04 s — first lick after odour onset, the bound that makes the wave pre-motor Quotation withheld because this record is not cleared for redistribution. 82° and 84°, and 65° — angles between the three activity axes before orthogonalisation was imposed Quotation withheld because this record is not cleared for redistribution. ~50% — trial-averaged task-related variance captured by the three modes together Quotation withheld because this record is not cleared for redistribution. 97 ± 1% versus 94 ± 2%, P = 0.67 — hit rate thirsty against stimulated, the null supporting behavioural restoration Quotation withheld because this record is not cleared for redistribution. 1195 of 1781,67% — state-modulated neurons that reversed their satiety change under stimulation Quotation withheld because this record is not cleared for redistribution. 18/29 regions — regions with significantly reduced odour decoding when sated 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.aav3932.