From neurotransmitters to dynamic connectivity: A statistical mechanics approach to modelling cortical interactions.

人类大脑不断地重组其连接，以应对我们每天经历的事件。我们改变大脑连接的能力被称为神经可塑性，它奠定了我们学习和发展的能力，以及在经历身体或心理创伤或伤害后康复的能力。关于可塑性背后的确切生物学机制，我们仍然一无所知。然而，清楚地了解可塑性是如何发挥作用的，将对临床神经科学产生重大影响。进一步了解健康的大脑是如何发育和衰老的，这将直接造福于那些患有某种脑损伤的人，他们试图刺激可塑性以求恢复，或者只是为了防止衰退或提高认知能力而刺激可塑性。目前，临床医生正在使用许多不同的技术来刺激可塑性。从本质上讲，大多数技术都涉及对目标大脑区域的电、磁或感觉刺激。如果神经元在同一时间被多次激活，那么它们之间的联系就会变得更强。相反，如果神经元没有被使用，那么它们的连接就会失去力量，最终停止所有的工作。这是希伯来学习和可塑性的基础。同时刺激大脑区是一种激活大量神经元的方式，从而使网络中的连接强度增强。然而，由于我们并不完全了解网络可塑性变化背后的机制，我们无法优化这些方法，使人们能够从治疗中获得最大利益。将数学和计算技术应用于生物问题是探索这些复杂系统如何运行的一个非常强大的工具。大脑现在被认为是一个高度动态和复杂的器官，为应对其复杂性而设计的数据分析方法正在不断发展。在单个细胞的微观水平上的数学建模工作提供了对一个细胞内或少数细胞之间潜在的化学反应的生物过程的深刻洞察。到目前为止，这些模型还没有扩大到包括跨越整个大脑区域的网络交互作用，以响应不同类型的刺激。我们最近开发了一个大脑可塑性模型，能够预测人们对两种不同刺激做出反应的行为变化。然而，该模型无法解释大脑内发生的化学变化，因此我们无法理解这些变化与行为观察的关系。我建议开发一个数学模型，能够解释大脑内活动、连接和化学物质浓度随着不同类型的外部刺激而发生的变化。研究模型对不同类型刺激的反应将使我能够预测大脑将如何对用于增强可塑性的疗法做出反应。

The human brain constantly reorganises its connections in response to the events that we experience every day. Our ability to modify the connections in our brain is called neuroplasticity and it underlies our capacity to learn and develop as well as to heal after experiencing physical or psychological trauma or injury. The precise biological mechanisms underlying plasticity still elude us. However, a clear understanding of how plasticity works will have great implications for clinical neuroscience. Furthering our knowledge of how healthy brains develop and age will directly benefit people who have suffered from a form of brain injury and are trying to stimulate plasticity in order to recover, or are simply stimulating plasticity to prevent decline or enhance cognitive performance.A number of different techniques are currently being used by clinicians to stimulate plasticity. Essentially, most techniques involve electrical, magnetic or sensory stimulation of a target brain region. If neurons are activated at the same time many times over, then the connections between them get stronger. Conversely, if neurons are not used then their connections lose strength and eventually stop working all together. This is the basis of Hebbian learning and plasticity. Stimulating large brain areas at the same time is a way of activating large populations of neurons so that the connection strengths within the network grow. However, as we do not fully understand the mechanisms underlying the changes we see in network plasticity, we are unable to optimise these methods so that people can get the maximum benefit from their treatment.Applying mathematical and computational techniques to biological problems is a very powerful tool for exploring how these complicated systems operate. The brain is now recognised as a highly dynamic and complex organ and data analysis methods designed to cope with its' complexity are constantly evolving. Mathematical modelling work at the microscopic level of individual cells has provided great insight into the biological processes underlying chemical reactions within a cell or between small numbers of cells. So far, these models have not been scaled up to include network interactions spanning over whole brain regions in response to different kinds of stimulation. We have recently developed a model of plasticity in the brain that was able to predict behavioural changes in people responding to two different kinds of stimulation. However, the model could not account for the chemical changes happening within the brain and so we were unable to understand how these were related to the behavioural observations. I am proposing to develop a mathematical model that is able to account for changes in activity, connectivity and chemical concentrations within the brain in response to different kinds of external stimuli. Studying the response of the model to different kinds of stimulation will allow me to to make predictions about how the brain will react to therapies used to boost plasticity.