Direct Cortical Stimulation to Manipulate Human Memory Function (MEMSTIM)

我们对过去事件的记忆定义了我们是谁，并允许我们想象和计划未来。因此，深入了解人类记忆功能的基本机制具有广泛的意义，从教育，目击者证词和智能机器的设计，到与压力，创伤和痴呆症相关的记忆功能障碍的治疗。海马体的损伤会对记忆产生深远的影响，在严重的情况下，会导致完全无法记住新事件。在海马体内，单个“位置细胞”和“概念细胞”的活动为特定的位置、人、动物和物体编码。据信，在记忆痕迹被“巩固”在更广泛的大脑区域网络中以支持记忆引导的行为之前，正在进行的经验通过加强这些神经元之间的突触连接而被迅速捕获。至关重要的是，支持这些功能的神经机制仍然难以捉摸。来自啮齿动物的电生理学数据表明，巩固和规划可能支持海马神经元活动序列的“重放”，但人类的因果关系证据缺失。为了弥合这一解释性差距，我将记录海马单细胞活动在术前癫痫患者进行认知任务，探测事件记忆的编码，存储和回忆。此外，我将使用一种创新的皮层刺激范式来检测和破坏真实的时间内特定事件的记忆，寻找记忆引导行为的因果关系;以及最先进的MEG解码技术，将这些发现扩展到整个大脑水平。我将综合这些结果与现有的知识，从计算建模，啮齿动物电生理学和神经成像提供直接的经验支持，人类记忆功能的神经水平模型，链接在人脑中的单细胞的动态行为，照亮新的治疗途径记忆障碍。

Our memories of past events define who we are and allow us to imagine and plan for the future. Gaining insight into the fundamental mechanisms of human memory function therefore has implications ranging from education, eye witness testimony, and the design of intelligent machines, to the treatment of memory dysfunction associated with stress, trauma, and dementia. Damage to the hippocampus can have profound effects on memory which, in severe cases, results in a complete inability to remember new events. Within the hippocampus, the activity of single 'place cells' and 'concept cells' code for specific locations, people, animals, and objects. It is believed that ongoing experience is rapidly captured by the strengthening of synaptic connections between these neurons before memory traces are 'consolidated' within a wider network of brain areas to support memory guided behaviour. Crucially, the neural mechanisms supporting these functions remain elusive. Electrophysiology data from rodents suggest that both consolidation and planning may be supported by the 'replay' of activity sequences in hippocampal neurons, but causal evidence from humans is missing. To bridge this explanatory gap, I will record hippocampal single cell activity in pre-surgical epilepsy patients performing cognitive tasks that probe the encoding, storage, and recall of event memories. In addition, I will use an innovative cortical stimulation paradigm to detect and disrupt memory for specific events in real time, looking for a causal role in memory guided behaviour; and state of the art MEG decoding techniques to extend these findings to the whole brain level. I will synthesise these results with existing knowledge from computational modelling, rodent electrophysiology and neuroimaging to provide direct empirical support for a neural level model of human memory function that links the dynamics of single cells in the human brain to behaviour, illuminating novel treatment pathways for memory disorders.