Ventral striatal processing of prefrontal inputs and phasic dopamine during rule switching

在新奇的情况下，个体通常通过反馈来学习，而反馈则预示着奖励。每当规则发生变化时，都需要更新有关操作及其后果的信息。规则转换和相关的大脑回路在哺乳动物中是保守的。这些脑回路包括腹侧纹状体的伏隔核(NAC)和内侧前额叶皮质(MPFC)。多巴胺(DA)瞬变既能预测奖赏，又能抑制mPFC和NAC的相互作用。我们将重点关注NAC，它被提议作为中央门户，在那里融合投入有助于优化奖励结果。然而，规则转换的神经机制在很大程度上是未知的。为了研究自由活动大鼠规则转换的神经关联，我们建议在执行策略集转换范式的操作版本期间，在NAC和mPFC中进行四极管记录和快速扫描循环伏安。实验将以两个假设为指导。首先，在NAC中集成一种新规则的mPFC表示，以优化奖励结果。我们预测，当动物还没有学习到新的规则时，规则转换后mPFC-NAC的相互作用会暂时增加。为了确定mPFC和NAC的相互作用，我们将研究神经元组件的协调放电及其由振荡引起的时间组织。其次，预测奖励的暗示会引发DA瞬变。我们假设，在规则改变后，旧的线索不再预测奖励，因此不再引发DA瞬变。因此，人为地保持DA信号与旧线索同时进行将防止行为规则切换。这些人工DA瞬变也被预测为抑制mPFC-NAC相互作用。为了达到这些目标，我们将初步研究在规则转换过程中，NAC中的DA相变与NAC中的行为表现和神经元活动的关系。在第二组实验中，在规则转换之后，将利用对DAR能神经元的光遗传刺激，在现在没有奖励的线索的同时，人工恢复DA瞬变。在此操作期间，我们将跟踪行为性能和mPFC-NAC交互。总之，我们的项目将有助于理解DA瞬变在规则切换和相关的NAC表示中的作用。这些发现对认知灵活性有影响，认知灵活性在主要神经精神障碍中受损。

In novel situations individuals often learn through feedback which cues predict reward. Information about actions and their consequences needs to be updated whenever rules switch. Rule switching and the involved brain circuits are conserved across mammals. These brain circuits contain the nucleus accumbens (NAc) of the ventral striatum and the medial prefrontal cortex (mPFC). Dopamine (DA) transients both predict rewards and suppress interactions of mPFC and NAc. We will focus on the NAc that has been proposed to serve as a central gateway where converging inputs help to optimize reward outcome. Yet, the neuronal mechanisms of rule switching are largely unknown. To study the neural correlates of rule switching in freely moving rats, we propose to perform tetrode recordings and fast scan cyclic voltammetry in NAc and mPFC during performance of an operant version of a strategy set shifting paradigm. Experiments will be guided by two hypotheses. First, mPFC representations of a novel rule are integrated in the NAc to optimize reward outcome. We predict that mPFC-NAc interactions transiently increase after the rule switch when the animal has not yet learned the novel rule. To determine interactions of mPFC and NAc, coordinated firing of neuronal assemblies and their temporal organization by oscillations will be examined. Second, cues that predict reward elicit DA transients. We hypothesize that after the rule has switched, the old cue does not predict reward anymore and therefore stops eliciting DA transients. Artificially maintaining the DA signal concurrent with the old cue will thus prevent behavioral rule switches. These artificial DA transients are also predicted to suppress mPFC-NAc interactions. Towards these aims, we will initially examine the relation of the phasic DA transients in the NAc to behavioral performance and neuronal activity in the NAc during rule switching. In a second set of experiments, after the rule switch, the DA transient will be artificially restored concurrent with the now unrewarded cue using optogenetic stimulation of DAergic neurons. During this manipulation we will track behavioral performance and mPFC-NAc interactions. Together, our project shall help to understand the role of DA transients in rule switches and associated NAc representations. These findings have implications for cognitive flexibility that is impaired in major neuropsychiatric disorders.