169 / 218
167
General discussion
processing signals about reward, cognition and action).
Combined, the experiments presented in this thesis confirm that striatal dopamine plays a
role in (motivated) cognitive control (
chapter 3
), and extend these findings by showing that
dopamine and the striatum are causally involved when information about rewards affects
cognitive goals (
chapter 4 and 6
). We also show that aberrant integration between reward
and cognitive control can surface in populations with deficits in cognitive control and striatal
dopamine signalling (
chapter 4 and 5
), suggesting that these cognitive deficits may stem from
deficits in translating information about potential rewards into cognitive goals. Dopamine
acts by changing the sensitivity of the neurons in the striatum, which is particularly sensitive
to input from the prefrontal cortex. We show that the prefrontal cortex modulates processing
in the striatum. We also show that signals about rewards in the reward-related corticostriatal
circuit are integrated with signals in the motor corticostriatal circuit when information about
reward, cognitive control and action control are integrated (
chapter 7
).
Which neural mechanism underlies the integration between reward and
cognitive control?
It is clear from the work in this thesis that the striatum and dopamine are crucial for motivated
cognitive control. It is also evident that top-down effects from the prefrontal cortex are
important for motivated cognitive control and that information from the anterior prefrontal
cortex during reward processing can affect processing in the motor part of the striatum (i.e. the
putamen). Anatomical evidence from non-human primates has revealed three ways in which
information can be transferred between circuits (
chapter 1 and 2
): Via direct cortico-cortical
(CC) connections, via cortico-striatal (CS) connections or via striato-nigral-striatal (SNS)
connections (
figure 1.1 and 2.1
) (Haber et al., 2000; Wood and Grafman, 2003; Haber and
Knutson, 2010). The work in this thesis clearly shows a role for the striatum during integration
of information across task-related goals (
chapter 4, 6, 7
), suggesting that information transfer
takes place at the level of the striatum. Although the work in this thesis strengthens the case for
a role for CS or SNS connections, it is important to realize that this does not rule out entirely
that the integration (also) takes place via CC connections. It may still be the case that the
striatum receives input from the cortex after signals have been integrated at the cortical level.
Given the results of the work presented here I would however speculate that the integration
takes place either by means of direct CS connections and/or by means of SNS connections. In
the first case (CS connections), the prefrontal input to the striatum is modulated by dopamine
(originating from the midbrain). In the case of SNS connections, dopaminergic connections
from the midbrain convey information to increasingly posterior/dorsolateral regions of the
striatum, which in turn project back to the midbrain (and the prefrontal cortex). The work
is
chapter 6
showed that lesions of the ventral striatum clearly affected motivated cognitive
control. This argues against a sole role for direct cortico-striatal connections, i.e. connections
from reward-related cortical regions to the cognitive striatum, bypassing the ventral striatum.