Dopamine agonist-induced impulse control behaviors: impulsivity, prediction error and risk.
Doctoral thesis, UCL (University College London).
A range of pathological behaviours associated with treatment with dopamine receptor agonists (DA) have been reported in Parkinson's disease (PD). The behaviours, known as impulse control disorders (ICD), are linked by repetitive rewarding or motivating choices despite negative consequences and include pathological gambling, compulsive shopping, binge eating and hypersexuality. This patient group provides a model for the study of the effect of both chronic DA and the interaction between DA and vulnerability mediated by cognitive mechanisms. In this thesis, I focus on PD patients with problem gambling, or compulsive shopping behaviors compared to matched PD controls without ICDs and matched healthy volunteers. I begin by outlining the methods used in this study including fMRI, reinforcement learning and model-based fMRI. I then ask whether these behaviours are associated with enhanced learning from gain feedback or impaired learning from loss feedback. As exogenous dopaminergic medications may interfere with the physiological function of phasic dopaminergic activity as a teaching signal, I use a reinforcement learning model to examine the effects of DA and the interaction of DA and vulnerability on prediction error activity as measured in the fMRI. Here, positive reinforcement is simply defined as “any event that increases the probability of a response” and does not imply hedonic value. I show a dissociation as a function of vulnerability: dopamine agonists are associated with faster reward learning in ICD patients and slower loss learning in PD controls. The effects appear to be mediated by differences in striatal prediction error activity. I then examine the effect of DA on various forms of impulsivity including impulsive choice (or the ability to delay gratification) and reaction time with increasing decision difficulty or decision conflict. I show here that DA in ICD patients are associated with greater impulsive choice and more rapid decisions compared to off DA and PD controls. ICD patients overall have faster reaction times compared to PD controls. I also show that ICD patients on DA have worse working memory function compared to PD controls on DA. I then examine the neural correlates of the impulsive choice task in healthy volunteers in an fMRI study. In the final data chapter, I assess the effects of DA on risk taking. I show that DA increases the tendency towards risk taking choices in ICD patients accompanied by a decrease in ventral striatal activity to an index of risk. Thus, DA may lead to an underestimation of risk in ICD patients by decreasing the perception of the difference between the potential for gain and the potential for loss outcomes in the risky choice. Thus, I construct a model of the ICD behavior. In a vulnerable subset of the PD population (presumably reflecting genetic, biological or psychosocial factors), DA hastens learning from rewarding outcomes emphasizing the acquisition phase of these behaviors and DA decreases the estimation of risk. DA appears to increase sensitivity towards immediate rewards over delayed rewards, an effect that may be mediated by the influence of DA on reward magnitude. These cognitive mechanisms can help explain why ICD patients choose to pursue these short term potentially rewarding risky gains despite marked negative financial losses and social and occupational consequences.
|Title:||Dopamine agonist-induced impulse control behaviors: impulsivity, prediction error and risk|
|Additional information:||Permission for digitisation not received|
|UCL classification:||UCL > School of Life and Medical Sciences > Faculty of Brain Sciences > Institute of Neurology|
Archive Staff Only