Deciding 'if' and 'when' to act: A neuroanatomical and neurochemical investigation

在进行自愿行动时，人们不仅要决定选择哪种行动，而且要考虑到在特定环境中采取行动的潜在好处，首先决定是否采取行动。想象一下，一只豹子正在权衡是去追逐一只鹿，还是继续在树荫下休息，期待着更好的猎物最终会沿着而来。尽管追求一个选择需要付出努力和时间，远离其他追求，采取行动的动机需要考虑潜在的回报，环境中其他机会的分布以及行动的成本。如果鹿是花豹的首选食物，花豹可能会决定采取行动并开始狩猎，但如果其他机会目前很少，或者如果鹿需要很少的努力来狩猎，她也可能这样做。如果成本、收益或环境（即，然而，如果情境信息（contextual information）不存在，那么就不会启动任何行动，豹子可能会保持不活动状态，继续休息。我的主要目标是了解情境信息如何影响动物的行动意愿，以及它如何通过大脑回路施加这种影响。为了实现这个目标，我将使用多种方法：我将训练猕猴完成一项简单的基于计算机的任务。他们将有机会采取行动以获得潜在的回报，并应决定是否采取行动以获得机会的好处并承担其成本，或者不采取行动并等待未来的机会。当动物完成任务时，当前机会的好处和它发生的环境会发生变化。我将使用数学模型来预测动物在当前环境下的行动意愿。当动物在执行任务时，我将用功能性磁共振成像（fMRI）记录它们的大脑活动。fMRI是一种非侵入性成像方法，可以产生大脑的整体图像。使用这种方法，我将确定大脑区域，调解上下文信息和动物的行动意愿之间的关系。然而，fMRI是一种“相关”方法。为了证明这种行为是否是由所识别的大脑区域“引起”的，我将使用经颅超声（TUS）神经刺激。TUS神经刺激是一种新颖的非侵入性方法，即使在表面的某个深度也可以短暂地局部改变大脑活动。功能磁共振成像和TUS显示大脑中的哪些区域跟踪情境信息对行动意愿的影响。然而，为了获得更广泛的了解，还需要确定传递这些信息的化学信使。在该项目的第二部分，我将使用TUS将药理学药物递送到大脑中的特定目标，并研究它们对大脑和行为的影响。通过这样做，我将能够确定在调解情境信息对行动意愿的影响方面发挥重要作用的主要化学信使。理解行动意愿的大脑机制不仅对于建立对决策的全面理解至关重要，而且在这样做的过程中，我们还可以设法了解这种回路功能障碍可能引起的适应不良行为，例如在环境变化时未能采取适当行动。这种失败包括冷漠和冲动，这是许多精神和神经疾病中普遍存在的症状。此外，在我的项目中有多种机会改善动物福利。例如，使用每只动物作为自己的对照（重复测量设计）减少了使用的动物数量，而非侵入性技术（fMRI，TUS）的应用减少了伤害，并有助于将发现转化为人类。这项研究将在牛津大学生物医学科学大楼进行，该大楼是世界领先的非人类灵长类动物研究中心之一，并得到跨学科合作者网络的支持。

When performing a voluntary action, one has to decide not only which action to choose but whether or not to initiate an action in the first place, given the potential benefits of acting in a particular environment. Consider a leopard weighing up whether to go to the trouble of chasing a deer or continue resting under the shadow of a tree in the expectation that better prey will eventually come along. Insofar as pursuing an option requires effort and time away from other pursuits, motivation to act requires considering the potential reward(s), the distribution of other opportunities in the environment, and the cost of action. The leopard may decide to act and start a hunt if the deer is her preferred food, but she may also do so if other opportunities are currently few and far between or if the deer requires little effort to hunt. If the costs, benefits, or environment (i.e., contextual information) are otherwise, however, then no action may be initiated, and the leopard may remain inactive and continue resting.My key goal is to understand how contextual information influences animals' willingness-to-act and how it exerts this influence via brain circuits. To achieve this goal, I will use a combination of methods:I will train macaques on a simple computer-based task. They will be presented with opportunities to act for potential reward and should decide whether to act in order to receive the benefits of the opportunity and incur its costs or refrain from action and wait for future ones. The benefits of a current opportunity and the environment in which it occurs change as animals go through the task. I will use mathematical models to predict animals' willingness-to-act given the current context. While animals are performing the task, I will record their brain activity with functional magnetic resonance imaging (fMRI). fMRI is a non-invasive imaging method that can produce an overall picture of the brain. Using this method, I will identify brain areas that mediate the relationship between contextual information and animals' willingness-to-act. fMRI, however, is a 'correlational' method. To show whether the behaviour is 'caused' by the identified brain areas, I will use transcranial ultrasound (TUS) neurostimulation. TUS neurostimulation is a novel and non-invasive method that can transiently and focally alter brain activity even at some depth from the surface.fMRI and TUS show which areas in the brain track the effect of contextual information on willingness-to-act. However, to get a broader picture one also needs to identify the chemical messengers that transmit this information. In the second part of the project, I will use TUS to deliver pharmacological agents to specific targets in the brain and investigate their effect on the brain and behaviour. By doing so, I will be able to identify major chemical messengers that play important roles in mediating the influence of contextual information on willingness-to-act.Understanding the brain mechanisms of willingness-to-act is not only essential for building a comprehensive understanding of decision making, but in doing so, we can also seek to understand maladaptive behaviours that potentially arise from dysfunction in this circuitry such as failure to act appropriately when environments change. Such failures characterise apathy and impulsivity, symptoms prevalent in a number of psychiatric and neurological disorders. Additionally, there are multiple opportunities within my project to improve animal welfare. For example, using each animal as its own control (repeated-measures design) reduces the number of animals used and the application of non-invasive techniques (fMRI, TUS) reduces harm and facilitates translation of findings to humans. This research will be carried out at the Oxford University Biomedical Sciences Building, home to one of the world-leading centres for non-human primate research, with support from a cross-disciplinary network of collaborators.