Proefschrift_Holstein

Chapter 2

Aging and cognitive control Aging is accompanied by a range of cognitive deficits and diminished striatal processing which are, at least partly, due to changes in the dopamine system. These changes in the dopamine system occur gradually across the life span and start early in adulthood (Backman and Farde, 2001). In chapter 5 I explored how integration of reward with cognitive control information changes across the life span, from adolescence to senescence. The results presented in chapter 5 show that aging is indeed accompanied by diminished integration between reward and task switching: younger subjects showed a reward-based adaptation of cognitive control, whereas responding in older adults did not vary with changing reward conditions. In summary, the DAT1 -depencency and BOLD fMRI work in chapters 3 and 4 suggest that striatal dopamine is involved in motivated cognitive control. In addition, previous work has shown age-related reductions in striatal dopamine. In chapter 5 I observed age- related changes in motivation-cognition integration ( chapter 5 ). Together these results suggest a role for the striatum in mediating this interaction. However, changes in BOLD response ( chapter 4 and box 2.4 ), DAT1 genotype-dependent effects ( chapter 3 and 4 ) or the correlation between aging and motivated cognitive control ( chapter 5 ) do not provide evidence for a causal role. When one wants to assess whether a region is crucial for a given function, the consequences of manipulating this region or its associated circuit is required. To assess whether the striatum was indeed crucial for successful integration of reward and task-switching signals, I aimed to disrupt processing in the ventral striatum in rodents. One challenge was the absence of a suitable paradigm to measure this effect in rodents. Paradigms in rodents often assess whether they can learn to flexibly adapt their behaviour, based on trial- and-error learning (i.e. without the use of cues) (Ragozzino et al., 1999). Further, although rewards are generally used to reinforce behaviour, the amount of reward an animal anticipates is often not directly manipulated, at least not on tasks measuring behavioural flexibility. To overcome this issue, I first developed a rodent homologue of the rewarded task-switching paradigm ( chapter 6 ). Next, I applied excitotoxic lesions ( box 2.5 ) to the rodent striatum (i.e. the nucleus accumbens core) to assess whether it is crucial for optimal integration of reward information and cognitive processes ( chapter 6 ). The results in chapter 6 showed reward-related improvements in cognitive flexibility in animals with an intact striatum, but not in those with lesions of the striatum. Together the results so far are in line with the role for striatal dopamine in motivated cognitive control. However, in chapter 1 we hypothesized that communication between the prefrontal cortex and the striatum may also be important for motivated cognitive control. More specifically, we hypothesized that signals in the prefrontal cortex might control activity in the striatum in a top-down manner.

42