Molecularly targetable cell types in mouse visual cortex have distinguishable prediction error responses

Neuron, Journal Year: 2023, Volume and Issue: 111(18), P. 2918 - 2928.e8

Published: Sept. 1, 2023

Predictive processing postulates the existence of prediction error neurons in cortex. Neurons with both negative and positive response properties have been identified layer 2/3 visual cortex, but whether they correspond to transcriptionally defined subpopulations is unclear. Here we used activity-dependent, photoconvertible marker CaMPARI2 tag mouse cortex during stimuli behaviors designed evoke errors. We performed single-cell RNA-sequencing on these populations found that previously annotated Adamts2 Rrad transcriptional cell types were enriched when photolabeling drive or responses, respectively. Finally, validated results functionally by designing artificial promoters for use AAV vectors express genetically encoded calcium indicators. Thus, distinct can be targeted using exhibit distinguishable responses.

Language: Английский

---

Regulation of circuit organization and function through inhibitory synaptic plasticity

Trends in Neurosciences, Journal Year: 2022, Volume and Issue: 45(12), P. 884 - 898

Published: Oct. 28, 2022

Language: Английский

---

Neural learning rules for generating flexible predictions and computing the successor representation

eLife, Journal Year: 2023, Volume and Issue: 12

Published: March 16, 2023

The predictive nature of the hippocampus is thought to be useful for memory-guided cognitive behaviors. Inspired by reinforcement learning literature, this notion has been formalized as a map called successor representation (SR). SR captures number observations about hippocampal activity. However, algorithm does not provide neural mechanism how such representations arise. Here, we show dynamics recurrent network naturally calculate when synaptic weights match transition probability matrix. Interestingly, horizon can flexibly modulated simply changing gain. We derive simple, biologically plausible rules learn in network. test our model with realistic inputs and data recorded during random foraging. Taken together, results suggest that more accessible circuits than previously support broad range functions.

Language: Английский

---

Prediction-error signals in anterior cingulate cortex drive task-switching

Nature Communications, Journal Year: 2024, Volume and Issue: 15(1)

Published: Aug. 17, 2024

Task-switching is a fundamental cognitive ability that allows animals to update their knowledge of current rules or contexts. Detecting discrepancies between predicted and observed events essential for this process. However, little known about how the brain computes prediction-errors whether neural prediction-error signals are causally related task-switching behaviours. Here we trained mice use switch, in single trial, responding same stimuli using two distinct rules. Optogenetic silencing un-silencing, together with widefield two-photon calcium imaging revealed anterior cingulate cortex (ACC) was specifically required rapid task-switching, but only when it exhibited signals. These were projection-target dependent larger preceding successful behavioural transitions. An all-optical approach disinhibitory interneuron circuit computation. results reveal mechanism computing transitioning states.

Language: Английский

---

Mechanisms for survival: vagal control of goal-directed behavior

Trends in Cognitive Sciences, Journal Year: 2023, Volume and Issue: 28(3), P. 237 - 251

Published: Nov. 29, 2023

Language: Английский

---

Simple synaptic modulations implement diverse novelty computations

Cell Reports, Journal Year: 2024, Volume and Issue: 43(5), P. 114188 - 114188

Published: May 1, 2024

Detecting novelty is ethologically useful for an organism's survival. Recent experiments characterize how different types of over timescales from seconds to weeks are reflected in the activity excitatory and inhibitory neuron types. Here, we introduce a learning mechanism, familiarity-modulated synapses (FMSs), consisting multiplicative modulations dependent on presynaptic or pre/postsynaptic activity. With FMSs, network responses that encode emerge under unsupervised continual minimal connectivity constraints. Implementing FMSs within experimentally constrained model visual cortical circuit, demonstrate generalizability by simultaneously fitting absolute, contextual, omission effects. Our also reproduces functional diversity cell subpopulations, leading testable predictions about synaptic dynamics can produce both population-level heterogeneous individual signals. Altogether, our findings simple plasticity mechanisms circuit structure qualitatively distinct complex responses.

Language: Английский

---

TreeCN: Time Series Prediction With the Tree Convolutional Network for Traffic Prediction

IEEE Transactions on Intelligent Transportation Systems, Journal Year: 2023, Volume and Issue: 25(5), P. 3751 - 3766

Published: Oct. 27, 2023

The complexity of traffic scenarios, the spatial-temporal feature correlations pose higher challenges for prediction research. Traffic model is an essential method in this research field, primarily focusing on capturing features among nodes and their neighboring nodes. However, existing methods lack comprehensive consideration directional hierarchical They are mostly applicable to scenarios with random uniform distribution nodes, but not suitable more complex small-scale aggregation scenarios. Therefore, study proposes Tree Convolutional Network (TreeCN), a tree-based structure. data design TreeCN focus relationships represented by plane tree matrix constructed as spatial matrix. TreeCN, full convolution network, performs bottom-up structure complete task node capturing. In study, thoroughly compared statistical, machine learning, deep learning time series prediction. experimental results show that only well also exhibits outstanding effect distribution. Moreover, adheres principles Graph Networks (GCN) can further capture them. This expected make new handle improve accuracy.

Language: Английский

---

Constructing Biologically Constrained RNNs via Dale's Backprop and Topologically-Informed Pruning

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 13, 2025

A bstract Recurrent neural networks (RNNs) have emerged as a prominent tool for modeling cortical function, and yet their conventional architecture is lacking in physiological anatomical fidelity. In particular, these models often fail to incorporate two crucial biological constraints: i) Dale’s law, i.e., sign constraints that preserve the “type” of projections from individual neurons, ii) Structured connectivity motifs, highly sparse defined connections amongst various neuronal populations. Both are known impair learning performance artificial networks, especially when trained perform complicated tasks; but modern experimental methodologies allow us record diverse populations spanning multiple brain regions, using RNN study interactions without incorporating fundamental properties raises questions regarding validity insights gleaned them. To address concerns, our work develops methods let train RNNs which respect law whilst simultaneously maintaining specific pattern across entire network. We provide mathematical grounding guarantees approaches both types constraints, show empirically match any constraints. Finally, we demonstrate utility inferring multi-regional by training network reconstruct 2-photon calcium imaging data during visual behaviour mice, enforcing data-driven, cell-type between spread layers areas. doing so, find inferred model corroborate findings agreement with theory predictive coding, thus validating applicability methods.

Language: Английский

---

Uncertainty-modulated prediction errors in cortical microcircuits

Published: Jan. 22, 2025

Understanding the variability of environment is essential to function in everyday life. The brain must hence take uncertainty into account when updating its internal model world. basis for are prediction errors that arise from a difference between current and new sensory experiences. Although error neurons have been identified layer 2/3 diverse areas, how modulates these learning is, however, unclear. Here, we use normative approach derive should modulate postulate represent uncertainty-modulated (UPE). We further hypothesise circuit calculates UPE through subtractive divisive inhibition by different inhibitory cell types. By implementing calculation UPEs microcircuit model, show types can compute means variances stimulus distribution. With local activity-dependent plasticity rules, computations be learned context-dependently, allow upcoming stimuli their Finally, mechanism enables an organism optimise strategy via adaptive rates.

Language: Английский

---

Layer-specific control of inhibition by NDNF interneurons

Proceedings of the National Academy of Sciences, Journal Year: 2025, Volume and Issue: 122(4)

Published: Jan. 22, 2025

Neuronal processing of external sensory input is shaped by internally generated top–down information. In the neocortex, projections primarily target layer 1, which contains NDNF (neuron-derived neurotrophic factor)-expressing interneurons and dendrites pyramidal cells. Here, we investigate hypothesis that shape cortical computations in an unconventional, layer-specific way, exerting presynaptic inhibition on synapses 1 while leaving deeper layers unaffected. We first confirm experimentally auditory cortex, from somatostatin-expressing (SOM) onto neurons are indeed modulated ambient Gamma-aminobutyric acid (GABA). Shifting to a computational model, then show this mechanism introduces distinct mutual motif between synaptic outputs SOM interneurons. This can control way competition for dendritic cells different timescales. thereby information flow redistributing fast slow timescales gating sources inhibition.

Language: Английский

---