Proefschrift_Holstein

Chapter 7

the aPFC modulates neural responses in the striatum, but it did not induce any behavioral changes, which is not uncommon with offline TMS (van Schouwenburg et al., 2012; Tupak et al., 2013). However, the absence of a behavioral effect precludes us frommaking any claims as to whether stimulation of the aPFC and the subsequent effects on the striatum had beneficial or detrimental effects on functional processes. Finally, contrary to our expectations, there was no modulation of task-related processing after stimulation of the dlPFC or PMC. The dlPFC region we stimulated showed significant overlap with the main effect of task switching. Nevertheless, stimulation of the dlPFC site failed to alter neural processing as a function of task switching or as a function of the interaction between task switching and response switching. One explanation is that the region we stimulated is not crucial for task-related processing in corticostriatal circuits in our paradigm. In fact, a meta-analysis has suggested a role for a more posterior region of the PFC in task switching (i.e. the inferior frontal junction; x, y, z coordinates: -40, 4, 30) (Derrfuss et al., 2005), which overlaps with the peak in the PFC activated by our task-switching contrast, suggesting indeed that the region we stimulated was too dorsal to target the corticostriatal circuitry involved in task switching. The absence of an effect of PMC stimulation likely reflects the finding that the network activated during Response switching did not show any overlap with the stimulation site in the PMC. We may have therefore failed to stimulate the region involved in response switching, and as a result we did not observe task-related modulation of PMC stimulation in the striatum. The SNS account discussed above predicts that the spiraling SNS connections are organized in an ascending way. We set out to test this idea by showing that stimulation of the aPFC, but not of the dlPFC or PMC, would affect processing in the striatum as a function of reward. This is exactly what we observed, although due to the absence of any effects after stimulation of the dlPFC and PMC we cannot be confident that stimulation of the dlPFC and PMC was effective. Thus, the results clearly show that task-related integration can occur across corticostriatal circuits and that is occurs in a unidirectional manner, from anterior/ventral to posterior/dorsal parts of the striatum. However, we cannot rule out that stimulation of the dlPFC or PMC, had it been effective, could also modulate activity in more anterior/ventral parts of the striatum. The current study is the first to show functional interactions between corticostriatal circuits during the integration of task-related goals, by causally manipulating neuronal excitability. The results of this TMS study show that corticostriatal circuits communicate in order to facilitate the translation of information across goals or functional domains. Understanding exactly how cognitive goals and subsequent actions are informed by reward motivation is important when understanding the etiology of a number of neuropsychiatric disorders with deficits in integrating between these signals and/or deficits in corticostriatal circuits (for a review see Shepherd, 2013), such as attention deficit hyperactivity disorder (Aarts et al., 2015; Hong et al., 2015), schizophrenia (Morris et al., 2015), obsessive compulsive disorder (Graybiel and Rauch, 2000), and addiction (Belin and Everitt, 2008; Tang et al., 2015).

158