
Neural Perturbational Inference (NPI) is a data-driven framework that non-invasively maps the directional, strength, and excitatory/inhibitory properties of effective connectivity across the entire human brain.


coremsg

mapping-causal-brain-interactions-using-virtual-perturbations-of-a-surrogate-neural-network


Mapping Causal Brain Interactions Using Virtual Perturbations of a Surrogate Neural Network


title_rewrite


A biologically plausible model for learning complex, non-Markovian sequences in recurrent networks of structured neurons, using a fully local, always-on plasticity rule that enables efficient and robust sequence learning.


efficient-learning-of-sequences-in-structured-recurrent-networks-a-biologically-plausible-model-for-cortical-sequence-learning


Efficient Learning of Sequences in Structured Recurrent Networks: A Biologically Plausible Model for Cortical Sequence Learning



중추복합체 뉴런의 활동은 현재 자극뿐만 아니라 이전 활동 이력에 의해서도 결정되며, 이는 빠른 방향 변화에 대한 반응을 촉진하고 직선 비행 중 에너지 소비를 줄일 수 있다.


Many insects use the polarization pattern of the sky for spatial orientation. Since flying insects perform rapid maneuvers, including saccadic yaw turns which alternate with translational flight, they perceive highly dynamic polarization input to their navigation system. The tuning of compass-neurons in the central complex of insects, however, has been mostly investigated with polarized-light stimuli that rotated at slow and constant velocities, and thus were lacking these natural dynamics. Here we investigated the dynamic response properties of compass-neurons, using intracellular recordings in the central complex of bumblebees. We generated naturalistic stimuli by rotating a polarizer either according to a sequence of head orientations that have been reported from freely flying bumblebees, or at constant velocities between 30°/s and 1920°/s, spanning almost the entire range of naturally occurring rotation velocities. We found that compass neurons responded reliably across the entire range of the presented stimuli. In their responses, we observed a dependency on spiking history. We further investigated this dependency using a rate code model taking spiking history into account. Extending the model to a neuronal population with different polarization tuning, which mirrored the neuronal architecture of the central complex, suggests that spiking history has a directly impact on the overall population activity, which has two effects: First, it facilitates faster responses to stimulus changes during highly dynamic flight maneuvers, and increases sensitivity for course deviations during straight flight. Second, population activity during phases of constant polarization input is reduced, which might conserve energy during straight flight.

Significance Statement Many insects use the pattern of polarized light, which arises from scattering of sunlight, for spatial orientation. Neurons in the central complex, a brain structure, which governs spatial orientation, encode the orientation of the insect’s head with respect to polarized light. To investigate the dynamic properties of these neurons, we recorded intracellularly in the central complex of bumblebees and stimulated with naturalistic polarized-light stimuli. We found that neuronal activity was not only dictated by the angle of polarization, but also by the amount of previous activity. Using a modelling approach we show, that this dependency on spiking history can facilitate faster adjustment of the heading signal and leads to reduced overall activity and therefore reduced energy consumption during straight flight.

### Competing Interest Statement

The authors have declared no competing interest.

History-dependent spiking facilitates efficient encoding of polarization angles in neurons of the central complex

자연스러운-편광각-자극에-대한-중추복합체-뉴런의-효율적인-부호화


자연스러운 편광각 자극에 대한 중추복합체 뉴런의 효율적인 부호화



Spiking history of polarization-sensitive neurons in the central complex of bumblebees facilitates faster responses to dynamic changes in polarization angles and reduces overall energy consumption during straight flight.


spiking-history-shapes-dynamic-encoding-of-polarization-angles-in-central-complex-neurons-of-bumblebees


Spiking History Shapes Dynamic Encoding of Polarization Angles in Central Complex Neurons of Bumblebees



NeuroPath is a novel deep learning model that integrates the coupling between structural and functional brain connectivity to enhance the prediction of cognitive states and uncover new insights into the complex relationship between brain structure and function.


neuropath-a-deep-learning-model-for-uncovering-the-coupling-between-structural-and-functional-brain-connectivity


NeuroPath: A Deep Learning Model for Uncovering the Coupling Between Structural and Functional Brain Connectivity



A novel Fuzzy Attention Layer integrated within a Transformer Encoder model can effectively capture interpretable patterns of neural activity from fNIRS data to decode human social interaction behaviors, such as handholding.


decoding-human-interaction-behaviors-using-a-fuzzy-based-approach-with-functional-near-infrared-spectroscopy


Decoding Human Interaction Behaviors Using a Fuzzy-based Approach with Functional Near-Infrared Spectroscopy



A novel workflow for integrating differentiable brain simulation, gradient-based optimization, and biologically-constrained neural network models to enable accurate multi-scale modeling of brain structure and function.


leveraging-differentiable-brain-simulation-for-accurate-multi-scale-modeling-of-brain-structure-and-function


Leveraging Differentiable Brain Simulation for Accurate Multi-scale Modeling of Brain Structure and Function



自発的な神経ネットワーク活動中、ニューロンの発火は、興奮性入力と抑制性入力の急速な変動に伴って起こることが明らかになった。


Self-sustained recurrent activity in cortical networks is thought to be important for multiple crucial processes, including circuit development and homeostasis. Yet, the precise relationship between synaptic input patterns and the spiking output of individual neurons remains largely unresolved. Here, we developed, validated and applied a novel in vitro experimental platform and analytical procedures that provide – for individual neurons – simultaneous excitatory and inhibitory synaptic activity estimates during recurrent network activity. Our approach combines whole-network high-density microelectrode array (HD-MEA) recordings from rat neuronal cultures with patch clamping and enables a comprehensive mapping and characterization of active incoming connections to single postsynaptic neurons. We found that, during network states with excitation(E)-inhibition(I) balance, postsynaptic spiking often coincided with the maxima of fast fluctuations in the input E/I ratio. These spike-associated E/I ratio escalations were largely due to a rapid bidirectional change in synaptic inhibition that was modulated by the network-activity level. Our approach also uncovered the underlying circuit architecture and we show that individual neurons received a few key inhibitory connections – often from special hub neurons – that were instrumental in controlling postsynaptic spiking. Balanced network theory predicts dynamical regimes governed by small and rapid input fluctuation and featuring a fast neuronal responsiveness. Our findings – obtained in self-organized neuronal cultures – suggest that the emergence of these favorable regimes and associated network architectures is an inherent property of cortical networks in general.

### Competing Interest Statement

The authors have declared no competing interest.

Parallel reconstruction of the excitatory and inhibitory inputs received by single neurons reveals the synaptic basis of recurrent spiking

自発的な神経ネットワークの活動中における単一ニューロンの入出力特性の詳細な解析により-自発的な発火のシナプス基盤が明らかになった


自発的な神経ネットワークの活動中における単一ニューロンの入出力特性の詳細な解析により、自発的な発火のシナプス基盤が明らかになった



Postsynaptic spiking during spontaneous recurrent network activity is primarily driven by rapid and brief changes in the balance of excitatory and inhibitory synaptic inputs, with a key role for a few strongly connected inhibitory hub neurons.


detailed-characterization-of-synaptic-inputs-reveals-the-mechanisms-underlying-spontaneous-spiking-in-recurrent-neural-networks


Detailed Characterization of Synaptic Inputs Reveals the Mechanisms Underlying Spontaneous Spiking in Recurrent Neural Networks



人類海馬體和內嗅皮質神經元能夠自發地整合「什麼」和「何時」的信息,提取持久且可預測的人類經驗時間結構表徵。


Single-neuron recordings from intracranial electrodes inserted into human brains for clinical reasons suggest that the temporal structure of human experience is encoded in human hippocampal and entorhinal neurons.

Human hippocampal and entorhinal neurons encode the temporal structure of experience - Nature

人類海馬體和內嗅皮質神經元編碼經驗的時間結構