Anne Fleur Kortekaas-Rijlaarsdam

CHAPTER 5 98 and TD children on a reversal learning task, although adolescents with ADHD showed reduced medial prefrontal cortex activation during reinforcement learning, suggesting impaired neural processing of reinforcement, which is necessary for learning in ADHD. Deficient reinforcement learning in ADHD has been hypothesized to be related to a dopamine-related deficit, as dopamine is a key neuromodulator of the (operant) conditioning process underlying reinforcement learning (Schultz, Dayan, &Montague, 1997). More specifically, ADHD has been associated with low phasic dopamine and reduced neural activity in ventral fronto-striatal pathways important for reinforcement learning (Frank, 2005; Nigg & Casey, 2005; Sagvolden, Johansen, Aase, & Russell, 2005; Tripp & Wickens, 2009; Wise, 2004). Methylphenidate (MPH), the most common treatment for ADHD (Antshel et al., 2011), increases the availability of dopamine in the fronto-striatal areas of the brain by blocking dopamine and noradrenaline transporters (Pliszka, 2005; Seeman & Madras, 1998; Volkow et al., 2001; Volkow et al., 1998). In 2007, Frank and colleagues tested their computational model predicting that ADHD-related reductions in phasic dopamine signaling would result in suboptimal action selection following reward (Frank, 2005; Frank et al., 2007). Adults on- and off stimulant medication were compared with controls to test the hypothesis that MPH would improve reinforcement learning. A probabilistic learning task with both fully informative and probabilistic feedback was used (Frank, Seeberger, & O’reilly, 2004). MPH increased the rate of acquisition of stimulus-response associations and the ability to apply information from the learning phase to novel stimulus pairings. Another study, using an instrumental learning task with both fully informative and probabilistic feedback, also showed positive effects of MPH on speed and accuracy of reinforcement learning of children with ADHD, irrespective of the reward probability condition, especially with high doses of MPH (Luman et al., 2015). So far, to the best of our knowledge, no studies have investigated the effects of MPH on reversal learning in ADHD. More knowledge on effects of MPH on reinforcement learning in children with ADHD is important as this type of learning is constantly required both in school and at home. In the current study we tested the hypothesis that reinforcement learning is impaired in children with ADHD and that MPH ameliorates these impairments. We extended the study of Frank et al. (2007) to children with ADHD and investigated stimulus-response learning, generalization of this knowledge and reversal learning. Furthermore, we compared MPH to placebo rather than no mediation (Frank et al., 2007), which is important to reduce possible risk of bias due to lack of blinding. Based on the dopamine deficit hypothesis, we predicted that children with ADHD would be impaired in acquiring stimulus-response associations