Decoding ensemble dynamics from cortico-amygdalar circuits during social choice

项目摘要 社会关系是人类健康和生存的关键组成部分， 行为对许多精神疾病有重大影响。尽管社会的重要性 在健康方面，仍然没有FDA批准的针对社会认知和 行为社会环境是由动物可获得的社会刺激所定义的，是一个关键的中介 影响啮齿类动物的行为，并影响社会选择。然而，我们对神经元回路 编码社会背景和选择。在啮齿动物中，投射到杏仁核（ACC- BLA）和核蛋白（ACC-Nac）已被证明是不同方面所必需的 社会信息传递。然而，这些区域的神经活动如何编码社会背景， 选择是未知的。我们开发了一个社会选择范例， 一个小说或一个熟悉的/笼子里的老鼠。在这个新的范例中，老鼠一直表现出 偏好的社会目标，而不是一个新的对象，但他们表现出可变的个人偏见，在社会 在小说和同类小说之间做出选择。记录ACC-BLA的神经活动 和ACC-Nac电路在这种行为模式下将允许更细致的 理解社会选择背后的神经机制。为了更好地理解 我们将如何利用在此期间记录的活动和传统的社会行为范式， 最近开发的统计模型和推理算法，集群非线性动力学 将神经反应分解成一个数目不详的功能性亚组，称为功能性编码 单位（FEU）。我们还将应用深度学习工具进行行为分析， 神经和行为数据的建模。这将增强我们预测社会环境的能力， 丰富社会行为的编码。最后，考虑到3，4-亚甲二氧基的影响， MDMA对社会行为和共情的影响及其临床意义 创伤后应激障碍，我们假设MDMA配对的社会接触会使社会偏见， 选择我们的范式我们将把深度学习应用于我们范例的行为分析， 为了验证这个假设。通过这个K奖，我们将定义社会刺激是如何编码的 在ACC-BLA和ACC-Nac回路中以特定于上下文的方式进行社会选择，以及如何 MDMA使这些社会选择产生偏见。与此同时，强化指导，指导阅读， 结构化的课程将提高我的技能和工具包在计算建模和机器 神经和行为数据的基于学习的分析，以及行为药理学，设置 独立的舞台。

Project Summary Social relationships are a key component of human health and survival and impairments in social behavior have a major impact in many psychiatric conditions. Yet despite the importance of social context to health, there remains no FDA-approved medications that target social cognition and behavior. Social context is defined by the social stimuli available to an animal, is a key mediator of behavior in rodents, and impacts social choices. Yet little is known about how neuronal circuits encode social context and choice.In rodents, circuits in ACC that project to the amygdala (ACC- BLA) and Nucleus accumbens (ACC-Nac) have been shown to be necessary for different aspects of social information transfer. However, how neural activity in these regions encode social context and choice is not known. We developed a social choice paradigm in which mice choose access to a novel or a familiar/cagemate mouse. In this novel paradigm, mice consistently show preference for a social target over a novel object, but they show variable individual biases in social choice between a novel and cagemate conspecific. Recording neural activity from the ACC-BLA and ACC-Nac circuits during this behavioral paradigm will allow for a more nuanced understanding of the neural mechanisms underlying social choice. In order to better understand how activity recorded during this and traditional social behavioral paradigms we will leverage recently developed statistical models and inference algorithms to cluster nonlinear dynamical neural responses into an unspecified number of functional sub-groups called Functional Encoding Units (FEUs). We will also apply deep learning tools for behavioral analysis to engage in joint modeling of neural and behavioral data. This will enhance our ability to predict social context and enrich encoding of social behavior. Lastly, given the impact of 3,4-Methylenedioxy methamphetamine (MDMA) on social behavior and empathy and its recent clinical significance in post traumatic stress disorder, we hypothesize that MDMA paired social exposure will bias social choice in our paradigm. We will apply deep learning to behavioral analysis of our paradigm in order to test this hypothesis. Through this K-Award we will define how social stimuli are encoded in a context-specific manner within ACC-BLA and ACC-Nac circuits during social choice and how MDMA biases these social choices. In parallel, intensive mentoring, directed readings, and structured coursework will enhance my skills and toolkit in computational modeling and machine learning-based analysis of both neural and behavioral data, and behavioral pharmacology, setting the stage for independence.