Establishing a spatial map of dopamine reward prediction error computations and their function in distinct associative learning processes across the striatum: a methodological framework

项目摘要/摘要。多巴胺(DA)信号在纹状体，主要输入基底节 神经节对于工具性学习至关重要，工具性学习是一个涉及刺激、反应和结果的联系的过程。 DA功能障碍会导致强迫症、帕金森氏症等疾病的不同症状 疾病和成瘾，这些通常被归因于不同的工具性学习过程中的不平衡。 纹状体解剖上分离的亚区被认为支持刺激-结果(S-O)，刺激-结果- 反应(S-R)和反应-结果(R-O)的联系。此外，当背内侧纹状体(DMS) 对于灵活的目标导向行为，背外侧纹状体(DLS)支持自动的、结果- 独立的习惯行为。而多巴胺(DA)通常被认为编码奖赏预测错误 (RPE)，一种驱动联想学习的教学信号，研究表明，DA的释放动力学各不相同 具体取决于目标区域。然而，尚不清楚自然时空DA释放动力学如何支持 学习不同的刺激、反应和结果关联。这些差距阻碍了有针对性的发展 影响不同纹状体区域的多巴胺功能障碍的诊断和治疗。 这一拟议的项目将在理解功能和计算方面取得进展 空间变化的DA动力学在不同的联想学习过程中的意义。一种行为范式 这要求小鼠从依赖线索的S-R策略切换到基于最近的 行动和结果将实现跨时间尺度的行为策略分类。这一行为范式 将与一种新的多光纤光度法相结合，在整个过程中记录DA的释放动力学 当小鼠学习和更新不同的刺激、反应和结果意外情况时，纹状体的体积。 这种新的大比例尺、细胞类型的特定记录方法将被应用于建立不同DA的空间地图 RPE可以关联并适用于记录任何脑区的分布式细胞类型的特定动态 高时空分辨率。最后，将该方法与数字镜像设备(DMD)结合起来进行目标定位 从光到大，但空间上精确的纹状体区域，用于光遗传操作，模拟空间 天然DA释放动力学的尺度和分辨率。 该项目的完成将支持三个主要领域的实践和理论培训： 测试和分析、功能电路分析和技术开发。马克·豪博士(赞助商)将 提供神经回路和动力学活体分析方面的指导和培训。David Boas博士(共同赞助人)， 波士顿大学神经光子学中心主任将提供概念和 用于光学神经工程的技术，这将加强由NSF支持的培训 神经光子学国家研究培训计划。神经科学研究生项目(GPN) 波士顿大学将提供额外的培训，同时培养一个协作和跨学科的环境。

PROJECT SUMMARY/ABSTRACT. Dopamine (DA) signaling in the striatum, the main input to the basal ganglia, is critical for instrumental learning, a process involving associations of stimuli, responses, and outcomes. DA dysfunction results in diverse symptoms in disorders such as obsessive-compulsive disorder, Parkinson’s Disease, and addiction, which are often attributed to an imbalance in distinct instrumental learning processes. Anatomically segregated subregions of the striatum are thought to support stimulus-outcome (S-O), stimulus- response (S-R), and response-outcome (R-O) associations. Further, while the dorsomedial striatum (DMS) is necessary for flexible goal-directed behavior, the dorsolateral striatum (DLS) supports automatic, outcome- independent habitual behavior. While dopamine (DA) is typically thought to encode a reward prediction error (RPE), a teaching signal which drives associative learning, studies suggest that DA release dynamics vary depending on the target region. However, it is unknown how natural spatiotemporal DA release dynamics support learning distinct stimulus, response, and outcome associations. These gaps hinder the development of targeted diagnostics and treatments for dopamine-dysfunction affecting distinct striatum regions. This proposed project will make strides toward understanding the functional and computational significance of spatially varying DA dynamics in distinct associative learning processes. A behavioral paradigm which requires mice to switch from a cue-dependent S-R strategy to a cue-independent strategy based on recent actions and outcomes will enable classification of behavior strategy across timescales. This behavioral paradigm will be combined with a new multi optical fiber photometry method to record DA release dynamics throughout the volume of the striatum as mice learn and update distinct stimulus, response and outcome contingencies. This new large-scale, cell-type specific recording method will be applied to establish a spatial map of distinct DA RPE correlates and can be adapted to record distributed cell-type specific dynamics of any brain region with high spatiotemporal resolution. Finally, this method will be advanced with a digital mirror device (DMD) to target light to large, yet spatially precise, regions of the striatum for optogenetic manipulation which mimics the spatial scale and resolution of natural DA release dynamics. Completion of this project will support practical and theoretical training in three main areas: behavioral testing and analysis, functional circuit analysis, and technology development. Dr. Mark Howe (sponsor) will provide mentorship and training in in vivo analysis of neural circuits and dynamics. Dr. David Boas (co-sponsor), the director of the Neurophotonics Center at Boston University, will provide training in the concepts and techniques used for optical neuro-engineering, which will augment training supported by the NSF Neurophotonics National Research Traineeship Program. The Graduate Program for Neuroscience (GPN) at Boston University will provide additional training while fostering a collaborative and interdisciplinary environment.