Explaining Consciousness as Neural Dynamical Complexity

对意识的科学解释是21世纪世纪科学的重要目标。从快速发展的大脑成像技术中收集到了大量与意识相关的经验数据，并形成了解释意识各个方面的理论的开端。现在有理由相信，我们即将在理解与意识相关的特定类型的大脑活动和大脑任务的特殊性方面取得重大突破。意识似乎涉及不同大脑区域之间精确平衡的通信量。太多的叫喊，没有一个地区能继续处理它的具体分配的工作。太少，什么也没有，将被捆绑在一起，成为一个统一的广播在整个大脑。如果在信息的区域隔离和全球整合之间有一个谨慎的平衡，那么全球广播的内容将产生一种意识体验。本项目将试图通过描述大脑活动数学模型中的这种微妙特性来测量意识。它旨在通过单独分析大脑数据来预测一个人的意识水平，例如区分他们是完全清醒，困倦还是深度睡眠。该项目将利用最先进的数据，包括手术植入电极的“颅内”记录，出于医学原因，这些电极被放置在大脑表面或大脑深处。分析将基于检测大量神经元活动产生的电场的电极组的记录。计算机模拟将有助于识别电极接收到的信号中与意识相关的活动的特征。新的数学模型、测量方法和统计技术相结合，将确保我们能够根据数据的性质，而不是根据活动的随机波动或测量的不准确性，可靠地做出关于意识的推断。意识科学尤其令人兴奋，因为它所揭示的东西对我们在自然界中的地位以及我们对自我的理解有着深远的影响。在实践层面上，有一个可靠的意识水平的测量将是非常有吸引力的临床。在某些情况下，基于患者行为的传统意识评估是不可靠的。在发生严重事故或中风后，患者可能会处于完全没有反应的状态，但仍然有意识，无法交流。这项研究将有助于为患有这种意识障碍的患者提供诊断信息，并指导他们治疗的伦理决策。这项研究也将在精神病学中找到应用，因为一些精神疾病可以被认为是意识障碍。了解意识的复杂神经机制将为诊断精神疾病和治疗精神痛苦开辟新的途径，该项目将提供一种新的方式来看待“复杂”系统，该系统被广泛认为是由许多组成部分组成的任何实体，这些组成部分以整体大于部分之和的方式相互作用。因此，所开发的数学和统计工具将具有远远超出人脑研究的应用潜力，例如，在信息技术、交通控制、气候变化和金融等复杂系统出现的几个领域。

A scientific account of consciousness is a key objective for 21st century science. A large array of consciousness-relevant empirical data has been gathered from rapidly advancing brain imaging technology, and the beginnings of theories accounting for aspects of consciousness have been formed. There is now reason to believe that we are on the verge of a major breakthrough in our understanding of what is special about the particular types of brain activity and brain tasks that are associated with consciousness. It appears that consciousness involves a precisely balanced amount of communication between the different brain regions. Too much shouting and no region can get on with processing its specifically assigned job. Too little and nothing will be tied together into a unified broadcast across the whole brain. If there is a careful balance between regional segregation and global integration of information, then the globally broadcast content will give rise to a conscious experience.This project will attempt to measure consciousness by describing this subtle property in mathematical models of brain activity. It aims to predict a person's level of consciousness by analysing brain data alone, e.g. to distinguish between whether they are fully awake, drowsy, or in deep sleep. The project will take advantage of state-of-the-art data, including `intracranial' recordings from surgically implanted electrodes that have, for medical reasons, been placed either on the brain's surface, or deep inside the brain. Analyses will be based on recordings from groups of electrodes that detect electric fields generated by the activity of large populations of neurons. Computer simulations will help identify the signatures of consciousness-related activity in the signals picked up by electrodes. A combination of new mathematical models, measures and statistical techniques will ensure that inferences about consciousness are made reliably based on properties of the data, and not from random fluctuations in activity or by inaccuracies in measurement.Consciousness science is particularly exciting because what is uncovered has profound implications for our place in nature and for our understanding of our very selves. At the practical level, having a reliable measure of conscious level would be extremely attractive in the clinic. There are scenarios in which traditional assessments of consciousness based on patient behaviour are unreliable. After a serious accident or a stroke, patients can be left in a condition in which they are completely unresponsive, yet could still be conscious and unable to communicate. This research will help inform diagnoses of patients suffering from such a disorder of consciousness, and guide ethical decisions on their treatment. This research will also find applications in psychiatry, since some mental illnesses can be considered as disorders of consciousness. Understanding the complex neural mechanisms of consciousness will open up new avenues for diagnosing mental illness and for treatment of mental suffering.This project will provide a new way of looking at `complex' systems broadly conceived as any entity that consists of many components that interact in such a way that the whole is greater than the sum of the parts. The mathematical and statistical tools developed will therefore have potential for application far beyond the study of the human brain, for example, to information technology, traffic control, climate change and finance, to name just a few domains in which complex systems arise.