Kenneth D. Miller

Linear NeuroPaint evaluates a linear model for cross-session neural alignment and inpainting, finding it less accurate than nonlinear models but computationally efficient.

The paper shows that paradoxical responses in brain regions can arise from inter-regional interactions rather than local inhibition stabilization.

The paper develops a nonlinear theory to predict cortical responses to optogenetic perturbations and identifies a feature-tuned inhibition pathway involving SST interneurons.

The paper constructs stabilized supralinear network models of mouse primary visual cortex to analyze how inhibitory cell types dynamically influence context-dependent processing and circuit stability.

The paper analyzes how genetically-defined cell populations encode natural scenes versus static gratings, finding that inhibitory cells show consistent encoding across individuals, and introduces IIRS as a metric for coding capacity.

This paper introduces Generative Inference, a framework that uses feedback pathways in neural networks to unify pattern recognition and generation, enabling perceptual grouping and illusory percepts.

The paper develops an exact linear theory to predict cortical network responses to perturbations by analyzing space- and feature-dependent connectivity.

The paper develops a theoretical framework to explain how untuned modulatory inputs interact with tuned sensory inputs in neural networks, showing suppressive baseline effects and rate reshuffling under specific conditions.

The paper analyzes neural population activity during decision-making and finds that decision trajectories lie on a curved 2D manifold, with distinct dynamics for slow and fast decisions.

The paper shows adaptation increases discriminability of frequent stimuli in mouse V1 by reshaping population response geometry, despite reduced individual neuron responses.

The paper extends an affine model to explain stimulus-dependent shared variability in visual cortex areas V1 and V2, linking circuit mechanisms to statistical models.

The paper shows that a retinotopically organized stabilized supralinear network with specific horizontal connections explains both contrast-dependent gamma oscillations and surround suppression in macaque V1.

The paper investigates how optogenetic stimulation reshuffles firing rates in mouse and monkey V1 by analyzing neuronal responses and developing theoretical models.

The paper uses optogenetic stimulation to investigate cortical contributions to cross-orientation suppression and finds that cortical circuits provide limited suppression, with sublinear integration observed at higher firing rates.

The paper develops a circuit model showing how feedforward and feedback processing in mouse V1 explain three contextual effects in visual responses.

This study shows that the human-specific gene SRGAP2C increases cortico-cortical connectivity and improves sensory processing and behavioral performance in mice.

The paper models how biased category representations in LIP neurons arise from interactions between Hebbian plasticity and recurrent connectivity during learning.

This paper introduces emergent property inference (EPI), a method using deep learning to solve inverse problems in theoretical neuroscience by learning parameter distributions consistent with computational properties.

The paper shows how inhibitory subnetworks in mouse V1 cortex switch between cooperation and competition to adapt to varying input strength and size.

The paper develops a stabilized supralinear network model of V1 layer 2/3 that reproduces surround suppression, surround tuning, and feature-specific suppression phenomena.

The paper identifies a disinhibitory circuit in mouse visual cortex that modulates responses based on stimulus-surround similarity.

The paper investigates how optogenetic and behavioral inputs modulate V1 circuit responses, showing that specific interneurons stabilize the circuit and locomotion affects excitatory cells through direct drive.

The paper analyzes paradoxical effects in excitatory/inhibitory neuronal networks, showing conditions under which cell types exhibit paradoxical responses based on network stability and connectivity.

This paper shows a circuit model of attention explains both neural and behavioral effects of visual attention across different experimental paradigms.

The paper uses an artificial neural network to study why excitatory and inhibitory neurons differ in selectivity and connectivity, finding that these differences can be explained by a few factors.

The paper investigates how attentional modulation affects performance in a deep convolutional neural network model of the visual system, finding that attention based on higher-level optimization outperforms attention based on neural tuning.

The paper tests the stabilized supralinear network model in the MT area of macaque visual cortex, finding similar modulatory effects as observed in V1 for motion direction stimuli.

The paper shows that a stochastic Stabilized Supralinear Network model explains stimulus-induced changes in cortical variability and correlations.