Selective Attention and Reward Value in the Prefrontal Control of Choice

描述(由申请人提供)：奖励价值整合和选择性注意是相互作用、为决策提供信息的深度交叉过程。然而，由于对这两个过程的研究在历史上存在分歧，这些过程的神经生物学机制尚不清楚。这是一个重大的遗漏，因为决策缺陷往往伴随着人类注意力缺陷，这表明这两个过程共同的神经机制存在一些干扰。在这里，我们提出了一系列实验来阐明这些机制。我们引入了一种行为范式，可以同时测量选择过程中的注意力和价值。我们将确定价值和注意力如何影响与注意力和决策有关的前额叶区域的单个神经元活动：额叶眼场(FEF)。在FEF内，不同的、特征明确的神经元类别一起工作来选择视觉目标。这些类型的神经元反映了选择形成的不同阶段：它们包括视觉(输入)细胞、运动(输出)细胞和视觉运动细胞，它们反映了中间计算。这些细胞分布在已知的皮质和皮质下连接的皮质层上：不同的皮质层本身反映了输入、输出和中间的局部连接。我不清楚价值和注意力在这些不同的神经类、层和它们相关的处理阶段中扮演什么角色。因此，我们将首先使用来自已识别的皮质层的高产量多电极记录来确定哪些神经类别发出信号价值和注意力。下一首, 我们将评估和操纵FEF的潜在价值信息来源。首先，我们将研究额叶前部的执行控制和动作价值区域：背侧前扣带回皮质(DACC)。使用两个区域的同步多电极记录，我们将检查两个区域中的值信号之间的时间关系。我们还将通过检查逐个试验的信号和噪声相关性来描述区域之间的功能连接。然后，我们将通过使dACC失活并观察其对FEF中选择形成的影响来确定dACC是否向FEF发送价值信息。最后，我们将询问多巴胺(DA)如何影响FEF中的值信号。我们将通过局部输送选择性激动剂和拮抗剂来操纵多巴胺(DA)，同时我们将从FEF中的单个神经元进行记录。我们将观察由此产生的对神经活动以及对注意力和价值观的独立行为测量的影响。拟议的实验将阐明前额叶皮质如何将注意力引导到选择选项上，权衡潜在结果的价值，然后将这些正在进行和重叠的过程整合到一个选择中。

DESCRIPTION (provided by applicant): Reward value integration and selective attention are deeply intersecting processes that interact to inform decisions. However, because of the historical divide between studies on the two processes, the neurobiological mechanisms through which these processes are unclear. This is a significant omission because deficits in decision-making are often accompanied by deficits in attention in humans, suggesting some disturbance in the neural mechanisms common to both processes. Here, we propose a series of experiments that will elucidate those mechanisms. We introduce a behavioral paradigm that can simultaneously measure attention and value during choice. We will determine how value and attention contribute to single-neuron activity in a prefrontal region that has been implicated in attention and decision-making: the frontal eye fields (FEF). Within the FEF, distinct, well-characterized classes of neurons work together to choose visual targets. These classes of neurons reflect different stages in choice formation: they include visual (input) cells, motor (output) cells, and visuomotor cells, which reflect intermediary computations. These cells are distributed across cortical layers, which have known cortical and subcortical connectivity: the various cortical layers themselves reflect input, output, and intermediate, local connections. It i unclear what role value and attention play in these distinct neural classes, layers, and their associated processing stages. Therefore, we will first use high- yield multi-electrode recordings from identified cortical layers to determine which neural classes signal value and attention. Next, we will evaluate and manipulate potential sources of value information for the FEF. First, we will look at a prefrontal executive control and action-value region with strong anatomical connections to the FEF: the dorsal anterior cingulate cortex (dACC). Using simultaneous multi-electrode recordings in both areas, we will examine the temporal relationship between value signals in the two regions. We will also describe the functional connectivity between the regions through examining trial-by-trial signal and noise correlations. We will then determine whether dACC sends value information to the FEF through inactivating dACC and observing its effects on choice formation in the FEF. Finally, we will ask how dopamine (DA) affects value signals in the FEF. We will manipulate dopamine (DA) through local delivery of selective agonists and antagonists while we record from single neurons in the FEF. We will observe the resulting effects on neural activity and on the independent behavioral measures of attention and value. The proposed experiments will elucidate how the prefrontal cortex directs attention towards choice options, weighs the value of potential outcomes, and then integrates those ongoing and overlapping processes into a choice.