Alexandre Pouget

This paper introduces a neural model that uses low-dimensional representations of module reuse to accelerate learning, showing consistency with primate behavioral and neural data.

This study examines age-related changes in neural and behavioural variability during a decision-making task in mice, finding increased neural variability and reduced variability quenching in specific brain regions.

This paper introduces modular neural state-space models to implement the Bayesian Brain hypothesis, enabling learned and reusable world-model components for flexible cognition.

This paper improves neural decoding by exploiting correlations across trials and sessions, showing better accuracy for four behaviors in mice using interpretable models.

Mice infer stimulus-side priors from recent choices; prior signals are decodable pre-stimulus in 24.0% to 30.2% of 242 brain regions, including early visual and motor areas.

The paper presents a brain-wide neural activity map from 621,733 neurons across 139 mice performing a decision-making task, showing how neural activity encodes task variables across brain regions.

Aeon enables long-term study of ethological behaviours in mice with simultaneous brain activity recording and provides a unified data format for analysis.

This paper shows that dopamine neurons in mice use multiple time scales to encode reward prediction errors, suggesting cell-specific temporal discounting and implications for reinforcement learning algorithms.

The paper shows that mice with ASD-related mutations exhibit similar computational and neural anomalies in prior updating during decision-making.

This paper uses curriculum-based reinforcement learning to train a neural network to control a musculoskeletal hand model, revealing insights into motor control and synergy analysis.

The paper argues that Weber's law arises from natural sensory input statistics rather than neuronal noise, and explains log-normal tuning curves in vivo.

The paper develops a neural model that uses natural language instructions to perform novel tasks with 83% accuracy through 0-shot learning, showing how language shapes sensorimotor representations.

This study investigates how Drosophila larvae combine sensory signals through probabilistic inference, showing adaptation to signal reliability and modality-dependent mechanisms.

The paper describes a modular data architecture for a neuroscience collaboration, enabling data contribution, access, and analysis across multiple labs.

This study shows that network connectivity determines population code performance by influencing stimulus tuning and internally generated fluctuations in spiking neuron models.

The paper shows mice switch between multiple decision-making strategies during perceptual tasks, challenging the assumption of a single strategy.

This study shows that learning influences speed-accuracy tradeoffs in decision-making by combining reinforcement learning with sensory evidence accumulation in rats.

The paper develops a method to detect differential correlations in neural data and applies it to V1 recordings, finding that V1 contains information consistent with behavioral performance.

The paper derives a Bayesian learning rule for diffusion decision models that incorporates confidence and prior biases to improve decision-making under uncertainty.

The paper derives the optimal policy for multi-alternative decisions and shows how it can be implemented by a neural circuit with normalization and urgency signals.

The paper shows that the brain integrates multisensory inputs optimally using a simple linear combination of neural activity, consistent with an invariant linear probabilistic population code.

The paper infers decoding strategies for two cortical areas in macaque monkeys using a mathematical framework that combines neural activity and inactivation experiments to resolve conflicting conclusions about brain function.