Parkinson's disease is a serious neurological disorder in which the dopamine-producing cells in the brain gradually degenerate. Patients are treated with dopaminergic medication, but this increases the risk of impulse control disorders, such as uncontrolled gambling, shopping, or eating.
Using the unique and extensive dataset from the "Parkinson Op Maat" study at Radboudumc, we were able to model which brain systems are involved in these disorders. We demonstrated not only that the prefrontal cortex is more active in people with both Parkinson's disease and impulse control problems during medication use, but also that this is associated with a higher expected value for rewards. Furthermore, by applying a unique combination of machine learning and generative modeling, we were able to predict which patients are more likely to recover from impulse control disorders. Our findings suggest that targeting the neural circuits involved in reward processing could offer new therapeutic opportunities.