Clinical and Computational Studies of Dopamine Function in Schizophrenia

描述(申请人提供)：精神分裂症是一种高度致残性疾病，影响0.5%-1%的美国人。与包括幻觉和妄想在内的疾病的阳性症状相比，疾病的残疾与认知缺陷和阴性症状的相关性更强。阴性症状通常被定义为缺乏正常的功能，但产生这种缺乏的实际机制仍不清楚，从而扼杀了合理的治疗发展。这个应用程序的首要目标是通过严格测试一个关于精神分裂症患者的嗜睡/快感缺失起源的高度具体的假设，从根本上改变对阴性症状的理解。这一假设已经在一个计算模型中形式化，该模型表明，精神分裂症患者缺乏表现刺激和反应备选方案的积极期望值的能力，以及从令人厌恶的结果中学习的完整能力。当不确定不同选择的可能回报时，这种代表价值的不足也被认为导致对行为选择的探索减少。该项目使用一个行为实验计划，在从结果中学习和决策方面测试这一假设。此外，我们还将探讨这一缺陷与目前关于阴性症状原因的认知心理学模型之间的关系，以及这一缺陷对于通过使用强化来塑造行为的行为治疗方法预测成功结果的重要性。这种计算方法还导致了一个关于神经机制的高度具体的假设，这些神经机制与代表期望值的缺陷有关。为了解决这一假设，我们将利用脑电的时间分辨率来测试神经活动的异常是像预测的那样发生在决策时，还是像预期的那样发生在反馈传递时，如果患者无法在结果好于预期的情况下使用多巴胺系统发出积极预测错误的信号，这是可以预期的。行为学和神经生理学研究的目标都是提供对阴性症状的明确的、机械性的理解，并评估这种方法在当前治疗方法中的应用。因为奖励回路在哺乳动物物种中是高度保守的，所以应该有可能“反向翻译”到指导药物开发研究所需的动物模型。

DESCRIPTION (provided by applicant): Schizophrenia is a highly disabling illness that impacts 0.5-1% of Americans. The disability of the illness is more strongly associated with cognitive deficits and negative symptoms than it is with the positive symptoms of the illness including hallucinations and delusions. Negative symptoms are generally defined as the absence of normal function, but the actual mechanisms involved in generating this absence have remained unknown, thereby stifling rational treatment development. The overarching goal of this application is to fundamentally alter the understanding of negative symptoms by rigorously testing a highly specific hypothesis about the origins of avolition/anhedonia in people with schizophrenia. This hypothesis has been formalized in a computational model that suggests that people with schizophrenia have a deficit in the ability to represent the positive expected value of stimuli and response alternatives, coupled with an intact ability to learn from aversive outcomes. This deficit in representing value is also thought to lead to reduced exploration of behavioral alternatives when uncertain about the likely payoffs of different choices. The project uses a program of behavioral experiments to test this hypothesis in the areas of learning from outcomes and decision making. In addition, we will explore the relationship between this deficit and current cognitive psychological models of the causes of negative symptoms, as well as the importance of this deficit for the prediction of successful outcome from a behavioral treatment approach that uses reinforcement to shape behavior. This computational approach also leads to a highly specific hypothesis about the neural mechanisms that are implicated in a deficit in representing expected value. To address this hypothesis, we will take advantage of the temporal resolution of EEG to test whether abnormalities in neural activity occur at the time of decision, as predicted, or instead occur at the time of feedback delivery, as would be expected if patients were unable to use the dopamine system to signal positive prediction errors when outcomes are better than expected. The goal of both the behavioral and neurophysiological studies is to provide an explicit, mechanistic understanding of negative symptoms and evaluate the application of this approach to current treatment approaches. Because reward circuitry is highly conserved across mammalian species, it should be possible to "back translate" to the animal models needed to guide drug development research.