Richard Naud

The paper shows that GABAergic synaptic dynamics form a functional continuum aligned with genetic identities, spanning Pvalb to VIP neurons with Sst as an intermediate class.

This paper introduces a probabilistic framework to uncover burst and population codes in neural data, showing that analyzing firing rates alone obscures latent structures.

This study shows that trisomy 21 impairs cortical circuit development by disrupting synaptic connectivity and intrinsic membrane properties in neurons.

The paper develops a method to analyze brain-wide electrical fluctuations and finds modular organization in the mouse brain that changes with context.

The paper investigates the cellular mechanisms of BTSP-LTP in the hippocampus, showing that theta-oscillation-paced inputs and acetylcholine modulate Ca2+ plateau potentials for efficient place-cell coding.

The paper analyzes serotonergic neuron activity during a dynamic Pavlovian task to estimate reward history and its relation to behavior.

This paper introduces and evaluates Burstprop, a burst-dependent algorithm for spiking neural networks, showing its effectiveness in image classification and robustness to feedback connectivity and weight discretization.

The paper identifies functional subtypes of short-term synaptic dynamics in mouse and human data using clustering and supervised learning, finding similar subtypes across species.

This study shows dendritic input controls interspike interval dispersion in neurons through three operating regimes, validated with experimental data.

This study uses deep learning to infer synaptic plasticity rules from single-neuron spiketrains, finding that transformers outperform other models but caution against applying simulation-trained classifiers to real data.

This paper proposes a theory that serotonin neurons encode predictive value signals, explaining various serotonergic responses better than existing theories.

This paper adapts the Burstprop algorithm for training hierarchical SNNs on MNIST, showing improved performance over STDP but lower accuracy than BPTT.

The paper investigates synaptic eligibility traces in mice prefrontal cortex neurons, showing how calcium dynamics and molecular mechanisms enable delayed credit assignment for learning.

The paper uses an augmented integrate-and-fire model to show that the dorsal raphe network encodes both input intensity and its temporal derivative, with the derivative dominating during rapid input increases.

The paper identifies two distinct neural codes, burst fraction and event rate, that convey stimulus information and error signals during learning, with burst fraction showing stronger modulation and temporal alignment with behavioral accuracy.

The paper identifies a slow serotonergic recurrent inhibitory network in the dorsal raphe nucleus that generates non-linear dynamics and winner-take-all computations, affecting reward-conditioned behavior.

The paper investigates cell-type specific responses to associative learning in the primary motor cortex using in vivo imaging in mice, finding differential modifications in neuronal activity patterns.

The paper shows that homeostatic inhibitory plasticity in interneurons can self-organize a network state where spikes and bursts occur asynchronously and irregularly, increasing information encoded in bursts.

This paper proposes parallel and recurrent cascade models to capture sub-cellular neural computation and integrates them into artificial neural networks for complex tasks.

This study shows that burst-dependent synaptic plasticity enables coordination of learning in hierarchical circuits by solving complex tasks through simulations and mathematical analysis.

The paper introduces a linear-nonlinear model to capture synaptic dynamics, showing it is more adaptable than previous models and can efficiently characterize synaptic processing.

The paper investigates how spike-timing patterns contribute to information transmission despite unimodal interspike interval distributions, finding that burst coding can still be effective even without clear bursts.

The paper shows that perirhinal cortex input to neocortical layer 1 is critical for early memory formation through dendritic calcium elevation and burst firing in layer 5 pyramidal neurons.

The paper develops a synthetic likelihood method to estimate silent synapse fractions, showing it reduces bias and improves statistical power compared to existing methods.

The paper develops a computational method to estimate quantal parameters at single synapses using iGluSnFR imaging and statistical algorithms.

This paper introduces a computational model to study how demyelination affects axonal transmission and shows that spike timing jitter can indicate demyelination levels.

The paper shows that cortical neurons can communicate multiple independent signals using sparse bursts, optimizing information transmission.

The paper shows that noise in dendrites can enhance information transmission in neural populations by switching leadership between dendrites and soma based on input strength.