Selection of experience for memory by hippocampal sharp wave ripples

Science, Год журнала: 2024, Номер 383(6690), С. 1478 - 1483

Опубликована: Март 28, 2024

Experiences need to be tagged during learning for further consolidation. However, neurophysiological mechanisms that select experiences lasting memory are not known. By combining large-scale neural recordings in mice with dimensionality reduction techniques, we observed successive maze traversals were tracked by continuously drifting populations of neurons, providing neuronal signatures both places visited and events encountered. When the brain state changed reward consumption, sharp wave ripples (SPW-Rs) occurred on some trials, their specific spike content decoded trial blocks surrounded them. During postexperience sleep, SPW-Rs continued replay those reactivated most frequently waking SPW-Rs. Replay awake may thus provide a tagging mechanism aspects experience preserved consolidated future use.

Язык: Английский

---

Abstract representations emerge in human hippocampal neurons during inference

Nature, Год журнала: 2024, Номер 632(8026), С. 841 - 849

Опубликована: Авг. 14, 2024

Humans have the remarkable cognitive capacity to rapidly adapt changing environments. Central this is ability form high-level, abstract representations that take advantage of regularities in world support generalization1. However, little known about how these are encoded populations neurons, they emerge through learning and relate behaviour2,3. Here we characterized representational geometry neurons (single units) recorded hippocampus, amygdala, medial frontal cortex ventral temporal neurosurgical patients performing an inferential reasoning task. We found only neural formed hippocampus simultaneously encode several task variables abstract, or disentangled, format. This uniquely observed after learn perform inference, consists disentangled directly observable discovered latent variables. Learning inference by trial error verbal instructions led formation hippocampal with similar geometric properties. The relation between format behaviour suggests geometries important for complex cognition. A which participants learned whose properties reflected structure task, indicating

Язык: Английский

---

Sleep microstructure organizes memory replay

Nature, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Язык: Английский

---

Learning produces an orthogonalized state machine in the hippocampus

Nature, Год журнала: 2025, Номер unknown

Опубликована: Фев. 12, 2025

Abstract Cognitive maps confer animals with flexible intelligence by representing spatial, temporal and abstract relationships that can be used to shape thought, planning behaviour. have been observed in the hippocampus 1 , but their algorithmic form learning mechanisms remain obscure. Here we large-scale, longitudinal two-photon calcium imaging record activity from thousands of neurons CA1 region while mice learned efficiently collect rewards two subtly different linear tracks virtual reality. Throughout learning, both animal behaviour hippocampal neural progressed through multiple stages, gradually revealing improved task representation mirrored behavioural efficiency. The process involved progressive decorrelations initially similar within across tracks, ultimately resulting orthogonalized representations resembling a state machine capturing inherent structure task. This decorrelation was driven individual acquiring task-state-specific responses (that is, ‘state cells’). Although various standard artificial networks did not naturally capture these dynamics, clone-structured causal graph, hidden Markov model variant, uniquely reproduced final states trajectory seen animals. cellular population dynamics constrain underlying cognitive map formation hippocampus, pointing inference as fundamental computational principle, implications for biological intelligence.

Язык: Английский

---

Space is a latent sequence: A theory of the hippocampus

Science Advances, Год журнала: 2024, Номер 10(31)

Опубликована: Июль 31, 2024

Fascinating phenomena such as landmark vector cells and splitter are frequently discovered in the hippocampus. Without a unifying principle, each experiment seemingly uncovers new anomalies or coding types. Here, we provide principle that mental representation of space is an emergent property latent higher-order sequence learning. Treating resolves numerous suggests place field mapping methodology interprets sequential neuronal responses Euclidean terms might itself be source anomalies. Our model, clone-structured causal graph (CSCG), employs scaffolding to learn representations by aliased egocentric sensory inputs unique contexts. Learning compress episodic experiences using CSCGs yields allocentric cognitive maps suitable for planning, introspection, consolidation, abstraction. By explicating role demonstrating how unify myriad observed phenomena, our work positions hippocampus sequence-centric paradigm, challenging prevailing space-centric view.

Язык: Английский

---

Sequential predictive learning is a unifying theory for hippocampal representation and replay

Опубликована: Апрель 29, 2024

Abstract The mammalian hippocampus contains a cognitive map that represents an animal’s position in the environment 1 and generates offline “replay” 2,3 for purposes of recall 4 , planning 5,6 forming long term memories 7 . Recently, it’s been found artificial neural networks trained to predict sensory inputs develop spatially tuned cells 8 aligning with predictive theories hippocampal function 9–11 However, whether learning can also account ability produce replay is unknown. Here, we find spatially-tuned cells, which robustly emerge from all forms learning, do not guarantee presence generate replay. Offline simulations only emerged used recurrent connections head-direction information multi-step observation sequences, promoted formation continuous attractor reflecting geometry environment. These trajectories were able show wake-like statistics, autonomously recently experienced locations, could be directed by virtual head direction signal. Further, make cyclical predictions future sequences rapidly learn produced sweeping representations positions reminiscent theta sweeps 12 results demonstrate how hippocampal-like representation engaged suggest reflect circuit implements data-efficient algorithm sequential learning. Together, this framework provides unifying theory functions hippocampal-inspired approaches intelligence.

Язык: Английский

---

A cellular basis for mapping behavioural structure

Nature, Год журнала: 2024, Номер unknown

Опубликована: Ноя. 6, 2024

Abstract To flexibly adapt to new situations, our brains must understand the regularities in world, as well those own patterns of behaviour. A wealth findings is beginning reveal algorithms that we use map outside world 1–6 . However, biological complex structured behaviours compose reach goals remain unknown. Here a neuronal implementation an algorithm for mapping abstract behavioural structure and transferring it scenarios. We trained mice on many tasks shared common (organizing sequence goals) but differed specific goal locations. The discovered underlying task structure, enabling zero-shot inferences first trial tasks. activity most neurons medial frontal cortex tiled progress goal, akin how place cells physical space. These ‘goal-progress cells’ generalized, stretching compressing their tiling accommodate different distances. By contrast, along overall was not encoded explicitly. Instead, subset goal-progress further tuned such individual fired with fixed lag from particular step. Together, these acted task-structured memory buffers, implementing instantaneously entire future steps, whose dynamics automatically computed appropriate action at each mirrored both on-task during offline sleep. Our suggest schemata structures can be generated by sculpting progress-to-goal tuning into buffers steps.

Язык: Английский

---

A persistent prefrontal reference frame across time and task rules

Nature Communications, Год журнала: 2024, Номер 15(1)

Опубликована: Март 8, 2024

Abstract Behavior can be remarkably consistent, even over extended time periods, yet whether this is reflected in stable or ‘drifting’ neuronal responses to task features remains controversial. Here, we find a persistently active ensemble of neurons the medial prefrontal cortex (mPFC) mice that reliably maintains trajectory-specific tuning several weeks while performing an olfaction-guided spatial memory task. This task-specific reference frame stabilized during learning, upon which repeatedly show little representational drift and maintain their across long pauses exposure repeated changes cue-target location pairings. These data thus suggest ‘core ensemble’ forming task-relevant space for performance consistent behavior periods time.

Язык: Английский

---

Dynamic prediction of goal location by coordinated representation of prefrontal-hippocampal theta sequences

Current Biology, Год журнала: 2024, Номер 34(9), С. 1866 - 1879.e6

Опубликована: Апрель 11, 2024

Язык: Английский

---

Prefrontal cortical ripples mediate top-down suppression of hippocampal reactivation during sleep memory consolidation

Current Biology, Год журнала: 2024, Номер 34(13), С. 2801 - 2811.e9

Опубликована: Июнь 3, 2024

Язык: Английский

---