Rafał Bogacz

The paper develops a framework to model how motivation affects choice and learning in the basal ganglia by integrating physiological states with reinforcement learning mechanisms.

The paper tests whether reward prediction errors (RPEs) influence risk-seeking behavior and finds that positive RPEs increase risk-seeking while negative RPEs increase risk-aversion.

The paper uses a Wilson-Cowan model to analyze phase-dependent effects of deep brain stimulation in essential tremor patients, showing the model can predict tremor responses and fit patient data.

This paper predicts the effects of deep brain stimulation using a reduced coupled oscillator model and proposes a hybrid strategy for closed-loop DBS in essential tremor patients.

The paper examines whether plasticity rules can extract payoffs and costs from reward prediction errors in basal ganglia pathways, showing they can infer action values with specific conditions.

The paper presents a theory explaining how vertebrate basal ganglia circuits select actions based on past experience and motivational state by integrating positive and negative consequences with neuromodulatory signals.

This paper shows that a predictive coding network with local Hebbian plasticity can approximate error back-propagation for supervised learning without external control.