NEURONAL BASIS OF PERSISTENCE

摘要 理解选择和控制的神经基础的传统方法涉及离散选择 任务在这些任务中，决策者在一个时间点做出明确的决策。虽然这些任务 虽然它们带来了宝贵的见解，但它们并不适合揭示自然的关键要素之一， 决定，其连续性。连续决策作为一种实验室工具， 研究持久性的神经基础，这是戒毒的关键因素， 通过离散的决策任务捕获。理解连续决策的神经基础至关重要 来理解决策的神经回路。使用离散选择任务， 神经科学在理解选择和控制方面已经取得了长足的进步，并开始确定选择和控制的作用。 背侧前扣带皮层（dACC）是一个关键的大脑结构。该区域单个神经元的活动 编码与离散决策的各个方面相关的多个变量， 这些决定的结果，并在事后适当调整行为。dACC的异常活动是 也与疾病的进展有关，这些疾病与自我控制和持久性差有关，例如 成瘾这些事实有力地表明，dACC在连续决策中也起着关键作用。我们 中心假设是，在连续决策过程中， 至少部分由dACC实现。研究连续决策的一个关键障碍是缺乏 易于使用的实验室任务，可在模型动物中训练，可可靠地进行数百次试验， 这会引起坚持，并激励不断调整。我们的实验室最近开发了一种 追捕恒河猴的任务在这项任务中，受试者使用操纵杆控制电脑上的一个化身 屏幕使用这个化身，他们追求虚拟逃跑的猎物，避免追求捕食者（换句话说，这是一个 类似于简化的吃豆人（Pac-man）。我们提出的研究计划将回答以下关键问题： dACC是否跟踪智能体在引导追踪中的运动？这些行为和神经基础是什么 未来的状态预测在引导追求？在持续的心理变化中， 决定？

Abstract Traditional approaches to understanding the neural bases of choice and control involve discrete choice tasks. In these tasks, decision-makers make categorical decisions at a single point in time. While such tasks have led to invaluable insights, they are poorly suited to shed light on one of the key elements of natural decisions, their continuous nature. Continuous decisions are especially interesting as a laboratory tool for studying the neural basis of persistence, which is a key ingredient of drug abstinence and which is difficult to capture with discrete decision-making tasks. Understanding the neural basis of continuous decisions is critical for developing an understanding of the neural circuitry of decision-making. Using discrete choice tasks, neuroscience has made great strides in understanding choice and control, and has begun to pin down the role of a key brain structure, the dorsal anterior cingulate cortex (dACC). Activity of single neurons in this region encodes multiple variables associated with aspects of discrete decisions and also with monitoring the outcomes of those decisions and adjusting behavior appropriately afterward. Aberrant activity in the dACC is also associated with the progression of diseases associated with poor self-control and persistence, such as addiction. These facts suggest strongly that the dACC plays a pivotal role in continuous decisions as well. Our central hypothesis is that both monitoring and decision processes in continuous decision-making are implemented, at least in part, by dACC. One critical barrier to studying continuous decisions is the lack of readily usable laboratory tasks that are trainable in model animals, reliably performable for hundreds of trials, that elicit persistence, and that motivate continuous adjustment. Our lab has recently developed a guided pursuit task for rhesus macaques. In this task, subjects use a joystick to control an avatar on a computer screen. Using this avatar, they pursue virtual fleeing prey and avoid pursuing predators (in other words, it is a kind of simplified Pac-man). Our proposed research program will answer the following key questions: How does the dACC track movements of agents in guided pursuit? What are the behavioral and neural bases of future state prediction in guided pursuit? What are the neural bases of changes of mind in continuous decisions?