Reward processing abnormalities have been observed in individuals with attention deficit/hyperactivity disorder (ADHD) in both behavioral and neuroimaging studies. Models of reinforcement learning...Show moreReward processing abnormalities have been observed in individuals with attention deficit/hyperactivity disorder (ADHD) in both behavioral and neuroimaging studies. Models of reinforcement learning in healthy individuals have laid the foundation for neurobiological theories addressing reward processing in ADHD. In healthy individuals, dopamine responses in ventral striatum (VS) gradually shift from actual rewards received (prediction error [PE]), toward cues which reliably predict such rewards (reward expectancy [RE]). Drawing on these observations, two theories posit that either low striatal dopamine (dynamic developmental theory) or failed signal shifts per se (dopamine transfer deficit theory) are behind reward-processing deficits in individuals with ADHD. However, the predicted signal abnormalities have not been examined directly. Forty-two participants with ADHD and 56 typically developing (TD) controls participated in a functional magnetic resonance imaging (fMRI) reward paradigm examining whole-task and temporal-change measures of PE and RE. Results showed that, contrary to theoretical predictions, the groups did not differ in either an overall measure of RE, or a composite index of PE-RE signal shifts. Furthermore, while overall PE activity was higher in the ADHD group (partly supporting the dopamine transfer deficit theory, which allows for high PE), observed decreases over time were similar between the groups (which was unexpected). Exploratory dimensional analyses showed that while a positive linear relationship between hyperactive/impulsive symptoms and RE was present in the full group, a quadratic (inverse U-shape) model better explained this relation in a sub-sample with currently-diagnosed ADHD, possibly supporting a model of downregulation due to higher symptoms. Finally, there were no significant associations between the index of temporal signal shifts and symptoms, or between overall PE and symptoms. In sum, results do not support the dynamic developmental theory, and only partially support the dopamine transfer deficit theory. Additionally, results suggest that overall signals, rather than dynamic changes, are better able to differentiate ADHD whether at the group level (PE) or at the individual level (RE). Increased PE in ADHD also suggests that immediate rewards (versus delayed reward anticipation cues) may be a useful strategy for interventions.Show less
