Back

Exploring reward learning disruptions as a possible mechanism underlying neuropsychiatric symptoms in Parkinson's disease

Sun, S.; Sharp, M.

2026-01-16 psychiatry and clinical psychology
10.64898/2026.01.14.26344132 medRxiv
Show abstract

BackgroundThe causes of neuropsychiatric symptoms in Parkinsons disease (PD) remain ill-defined. Disruptions in dopamine-dependent reward learning, a consequence of midbrain dopamine loss, potentially represent a mechanism that could underlie the neuropsychiatric symptoms of apathy, depression, and impulsivity, all of which have been proposed to reflect aberrant goal-directed behaviour. However no large-scale investigation that jointly considers these symptoms in PD has ever been undertaken. We aimed to determine if reward learning is associated with apathy, depression and impulsivity symptom in two samples of PD patients. MethodsTwo samples of PD patients (nsample1=81, nsample2=90), tested in their medicated state, completed two widely used reward learning tasks (probabilistic stimulus selection task, probabilistic reward task) from which we derived five summary measures of performance. Apathy, depression, and impulsivity were evaluated using validated self-report questionnaires. ResultsOverall, there was poor consistency in the relationship between reward learning and neuropsychiatric symptom severity. Greater depressive symptom severity was associated with slower reward learning performance in sample 1, but with both slower and faster reward learning performance in sample 2. Greater impulsivity symptom severity was associated with slower reward learning performance in sample 1 but not in sample 2. There were no associations between apathy and reward learning. ConclusionsWe found inconsistent relationships between symptoms of apathy, depression, and impulsivity and reward learning performance across two samples of PD patients. While this doesnt rule out the possibility that reward learning impairments contribute to these symptoms, it suggests any effect is likely to be small and overshadowed by other non-measured factors.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.