Timothy Verstynen

This paper develops a computational framework to analyze decision-making dynamics in the cortico-basal ganglia-thalamic network, revealing two distinct phases: deliberation and commitment, governed by different pathways.

The paper simulates learning in a spiking cortico-basal ganglia-thalamic model to show how subnetworks reshape decision dynamics through dopamine-dependent plasticity, shifting strategies from deliberation to commitment.

CBGTPy is a framework for simulating cortico-basal ganglia-thalamic dynamics in decision-making tasks, enabling testable predictions of behavior and neural activity.

This study investigates the role of arkypallidal neurons in the external globus pallidus in mediating inhibitory control by modulating competition in the striatum through a biologically-constrained spiking model.

The paper predicts instantaneous heart period from whole brain fMRI data using machine learning, identifying cortical and subcortical brain networks linked to psychological stress as reliable predictors.

The paper uses computational simulations to show how dopaminergic plasticity in cortico-striatal synapses adjusts control ensembles in CBGT circuits to improve reward rate during learning.

The paper shows how dopaminergic plasticity shifts evidence accumulation in humans by altering competition between action representations during decision-making.

The paper examines how D2 dopamine receptor expression and reward sensitivity affect learning in a complex decision-making task using human behavioral and neuroimaging data.

The paper compares task-related network profiles and activation patterns in two executive control tasks, finding greater differences in functional correlations than in evoked activity.

This study examines how humans adjust their decision policies in response to environmental changes by manipulating reward certainty and action reliability in a two-armed bandit task.

The paper investigates use-dependent learning (UDL) and finds that movement biases arise from both contaminated motor plans and execution biases, with planning dynamics showing a stronger influence.

The paper improves prediction of cognitive abilities by integrating multiple neuroimaging modalities using machine learning, achieving a 1-4% increase in variance explained compared to single-modality approaches.

The paper examines how structural brain network topologies relate to cognitive performance in a value-based decision-making task using diffusion MRI and graph theory measures.

This study shows that learning a movement sequence does not change the cortical representations of individual actions, despite behavioral evidence of binding.

The paper models how dopaminergic feedback alters corticostriatal pathway competition to influence evidence accumulation and decision-making.

The paper models human learning in a stop-signal task using reinforcement learning and decision models, identifying two distinct learning mechanisms for adapting action timing and inhibition.

The paper creates a population-averaged atlas of the human structural connectome using diffusion MRI data, revealing network topology features like small-worldness and hub structures.

The paper uses diffusion MRI to analyze white matter architecture and shows it can predict social, health, and cognitive attributes across individuals.

The paper shows that higher central adiposity is linked to slower learning rates in a sequential motor task, but not accuracy.

This paper introduces a method to quantify individual differences in whole-brain white matter architecture using local connectome fingerprints, achieving high accuracy in self-identification and revealing neuroplasticity and genetic influences.

The paper proposes that basal ganglia pathways dynamically compete to encode action uncertainty, linking neurocomputational theories of decision-making and reinforcement learning.

This study uses diffusion imaging to characterize the structural subcompartments of the human pallidum and their relation to nigrostriatal pathways.