Abstract
Dopamine systems mediate key aspects of reward learning. Parkinson’s disease (PD) represents a valuable model to study reward mechanisms because both the disease process and the anti-Parkinson medications influence dopamine neurotransmission. The aim of this pilot study was to investigate whether the level of levodopa differently modulates learning from positive and negative feedback and its electrophysiological correlate, the error related negativity (ERN), in PD. Ten PD patients and ten healthy participants performed a two-stage reinforcement learning task. In the Learning Phase, they had to learn the correct stimulus within a stimulus pair on the basis of a probabilistic positive or negative feedback. Three sets of stimulus pairs were used. In the Testing Phase, the participants were tested with novel combinations of the stimuli previously experienced to evaluate whether they learned more from positive or negative feedback. PD patients performed the task both ON- and OFF-levodopa in two separate sessions while they remained on stable therapy with dopamine agonists. The electroencephalogram (EEG) was recorded during the task. PD patients were less accurate in negative than positive learning both OFF- and ON-levodopa. In the OFF-levodopa state they were less accurate than controls in negative learning. PD patients had a smaller ERN amplitude OFF- than ON-levodopa only in negative learning. In the OFF-levodopa state they had a smaller ERN amplitude than controls in negative learning. We hypothesize that high tonic dopaminergic stimulation due to the dopamine agonist medication, combined to the low level of phasic dopamine due to the OFF-levodopa state, could prevent phasic “dopamine dips” indicated by the ERN needed for learning from negative feedback.
Highlights
The dopamine system plays a crucial role in reinforcement learning
Since we were interested in investigating whether dopaminergic medication differently affects learning from positive vs. negative feedback in Parkinson’s disease (PD), we focused the analysis on the accuracy of novel pair combinations in the Testing Phase and the error related negativity (ERN) associated to errors committed in these trials
Testing Phase—Within Patients Analysis Paired Samples T tests did not show any significant difference between PD patients ON-levodopa and PD patients OFFlevodopa in overall accuracy
Summary
The dopamine system plays a crucial role in reinforcement learning. In particular, electrophysiological studies in primates have shown that reward elicits phasic dopamine increases, whereas aversive feedback leads to phasic dopamine decreases in midbrain dopamine neurons. Moustafa et al (2013) proposed a reinforcement learning model that takes into account the different effects of levodopa and dopamine agonists on tonic and phasic dopamine levels, determining selective effects on reward learning processes. Levodopa enhances reward learning via phasic dopamine activity whereas dopamine agonists impair reward learning by increasing tonic dopamine activity, preventing phasic responses from being effective. These assumptions are substantiated by studies on healthy people reporting that administration of dopamine agonists negatively affects reinforcement-based learning (Frank and O’Reilly, 2006; Santesso et al, 2009)
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
