Dopaminergic circuitry underlying fear extinction learning

了解哪些刺激预示着危险的能力对生存至关重要，但当这些刺激不再构成威胁时，适应行为也同样重要。这方面的一个典型例子是恐惧消退学习，在此过程中，反复呈现不再预测厌恶结果（非条件刺激，US）的刺激（条件刺激，CS）会导致习得性恐惧反应逐渐减少。许多焦虑障碍，如创伤后应激障碍，其特征是恐惧消退受损，因此了解这种安全学习形式的神经机制具有很高的临床意义。恐惧消退代表了新的学习，其中CS与安全（美国的缺席）之间的联系是学习到的。这种联想学习是由预测误差（PE）驱动的，它表明预期结果与实际结果之间存在差异。为了启动消退学习，必须检测到预期的厌恶美国的缺失，并将其信号发送到调解恐惧消退的大脑区域。然而，这种PE信号的神经基质仍然是难以捉摸的。在灭绝期间没有收到预期的厌恶美国是一个好于预期的结果，可以作为一个惊喜和奖励事件来体验。因此，我们假设灭绝预测误差（EPE）可以被概念化为食欲性PE，因此可以由同样调节奖励学习的神经回路调节。与此一致，我们最近首次证明了位于腹侧被盖区（VTA）的多巴胺（DA）神经元子集在恐惧消退过程中编码EPE信号，并且该DA信号对于加速消退学习是必要的，也是足够的。然而，究竟如何以及通过下游目标，一个DAergic EPE信号启动和驱动恐惧消退目前尚不清楚。为了解决这个问题，我们将首先通过执行投影特异性活动依赖性钙记录来研究epe编码DA神经元的投影目标。投影特异性和时间精确的光遗传学操作，使用DA生物传感器测量DA释放以及药理学操作将进一步在确定的epe编码DA神经元的目标区域进行。确定了编码EPE的DA神经元的投射目标后，下一个目标将是研究这些神经元的独特的单突触输入，这些单突触输入可能为计算EPE信号提供信息。最后，为了进一步研究恐惧消退与奖励学习之间的关系，我们将研究发出EPE信号的DA神经元是否也编码奖励PE （RPE）。综上所述，拟议的实验将对多巴胺能EPE信号如何介导消退学习及其与奖励学习的关系产生重要见解。

The ability to learn which stimuli predict danger is crucial for survival but it is equally important to adapt behavior when those stimuli no longer represent a threat. One classic example of this is fear extinction learning, during which the repeated presentation of a stimulus (conditioned stimulus, CS) that no longer predicts an aversive outcome (unconditioned stimulus, US) leads to a gradual decrease in learned fear responses. Many anxiety disorders, such as post-traumatic stress disorder, are characterized by impaired fear extinction and thus understanding the neural mechanisms of this form of safety learning has high clinical significance. Fear extinction represents new learning in which the association between the CS and safety (the absence of the US) is learned. This kind of associative learning is driven by prediction errors (PE) that signal the discrepancy between expected and actual outcomes. In order to initiate extinction learning, the absence of the expected aversive US must be detected and signaled to the brain regions mediating fear extinction. However, the neural substrates of such a PE signal have remained elusive. Not receiving a predicted aversive US during extinction is a better-than-expected outcome and may be experienced as a pleasant surprise and a rewarding event. We therefore hypothesize that the extinction prediction error (EPE) could be conceptualized as an appetitive PE and could accordingly be mediated by the same neural circuitry that also mediates reward learning. Consistent with this, we have recently demonstrated for the first time that a subset of dopamine (DA) neurons, located in the ventral tegmental area (VTA), encode an EPE signal during fear extinction and that this DA signal is both necessary for, and is sufficient to accelerate, extinction learning. However, exactly how and through which downstream targets a DAergic EPE signal initiates and drives fear extinction is currently unknown. To address this question, we will first investigate the projection target of EPE-encoding DA neurons by performing projection-specific activity-dependent calcium recordings. Projection-specific and temporally-precise optogenetic manipulations, measurement of DA release using a DA biosensor as well as pharmacological manipulations will further be performed in the identified target region of EPE-encoding DA neurons. Having identified the projection target of EPE-encoding DA neurons, the next objective will be to investigate the unique monosynaptic inputs to these neurons that potentially provide the information for computing the EPE signal. Finally, in order to further examine the relationship between fear extinction and reward learning, we will investigate whether DA neurons that signal EPE also encode a reward PE (RPE). Taken together, the proposed experiments will yield crucial insights on how a dopaminergic EPE signal mediates extinction learning and its relationship to reward learning.