Proefschrift_Holstein

Chapter 5

accompanied by reward-based changes in flexible cognitive control, where older participants no longer exhibit an effect of reward motivation on task switching performance. Reward-induced changes in the speed-accuracy tradeoff have not been widely studied, and never in the context of a cognitive task. However, at least one study suggests that potential rewards can induce more cautious (i.e. more accurate and slow) behavior (Bijleveld et al., 2010). We replicate this effect in the cognitively less demanding repeat trials by showing that young participants exhibit reward-induced cautious behavior in the repeat condition. We extend this finding by showing that reward induced less accurate and faster responding on the more demanding switch trials in younger participants. These results suggest that reward can modulate cautious behavior differentially in distinct cognitive conditions. Reward motivation, cognitive control and the interaction between reward and cognitive control implicate, among other mechanisms, dopamine in the striatum (Roberts et al., 1994; Ikemoto and Panksepp, 1999; Aarts et al., 2010; Aarts et al., 2011). Furthermore, changes in speed-accuracy tradeoff strategies have been associated with changes in connectivity between the cortex and the striatum (Bogacz et al., 2010). More specifically, stronger corticostriatal connections have been found to promote faster (and premature) responses. The current study does not address the neural mechanisms of age-related effects on rewarded task switching directly. However, several independent studies have revealed age-related changes in dopamine signaling (Volkow et al., 1996b; Bäckman et al., 2000) Erixon-Lindroth et al., 2005), starting in early adulthood (Backman and Farde, 2005). Also, age-related decreases in corticostriatal connectivity, accompanied by slower (and more accurate) responses have been reported (Forstmann et al., 2011). Accordingly, here, we put forward the hypothesis that the observed change in speed-accuracy strategy during the integration of reward and cognitive performance reflects reduced dopamine signaling in the striatum, and a subsequent reduction in the adaptation of corticostriatal responses to the task conditions. The increased effect of reward in younger versus older participants is also in line with a number of previous findings in adolescents. First, neuroimaging work has revealed increases in reward sensitivity and ventral striatal responses in adolescents (Somerville and Casey, 2010). In addition, previous work has reported reward-related improvements in impulse control in adolescents (Kohls et al., 2009; Geier et al., 2010). In older participants, however, if anything a decrease in reward sensitivity is reported (Schott et al., 2007; Rademacher et al., 2014). We extend these findings by showing that young (i.e. < 26 years old) subjects can show reward-related adaptations of flexible cognitive control during task switching, whereas older participants (i.e. > 25 years old) cannot. One limitation of the current study is the fact that we pooled data from several studies. As a result, there are a number of factors of no interest that differ as a function of age, such as effects of maximum available reward. However, we accounted for this by replicating the effects in a subgroup of participants who received exactly the same amount of reward and number of trials. Nevertheless, it is clear that the reported results require replication in future, preferably longitudinal, work using a single study set-up. Also, it should be noted that a large number of

112