The role of cortico-hippocampal interaction in memory consolidation

系统记忆巩固是一个全脑范围的过程，它将最近编码的易记忆转化为更稳定和更长期的记忆。记忆过程的“两阶段”理论认为，在脱机状态下，记忆巩固在海马体和大脑皮层之间进行双向对话，海马体最初形成记忆痕迹，大脑皮层最终存储记忆痕迹，以便长期保留。在主动探索空间背景的过程中，海马CA1区(HPC)和初级视皮层(V1)的神经元亚群随着经验的发展而发展空间敏感性，从而这些空间调制的神经元将沿着动物的运行轨迹被顺序激活。值得注意的是，在离线期间，在两个区域观察到的活跃空间体验中的神经元序列可以以时空协调的方式被播放。这种重放抑制的中断将损害空间记忆性能。这些观察结果表明，在连贯重播前后，海马体和大脑皮层之间的相互作用可能有助于记忆巩固。然而，到目前为止，我们还不清楚重播前后皮质-海马区信息流的性质和方向。先前的证据表明，新皮质活动的变化可以先于重播事件。然而，目前尚不清楚重播事件之前的V1活动模式是否可以选择性地影响记忆内容在海马体和V1中重播。此外，我们不知道训练后回放过程中海马-皮质相互作用一致性的改善是否可以预测V1中学习到的空间信息的记忆性能和保真度。为了解决这些研究问题，我们打算从HPC的V1和CA1中记录，同时学习在虚拟现实(VR)中区分两个新的线性轨迹。在两个位置上具有视觉相同地标的两个VR线轨将被用来跟踪V1中空间调制随时间的发展。我们还将训练动物在到达每个线性轨道的指定奖励区时舔舔奖励，这鼓励动物主动区分两个线性轨道。通过记录几周来多个学习会话和学习后休息期间的V1和CA1神经活动，我们想要检查重放过程中的重放动力学和皮质-海马区相互作用是如何随着时间的推移而改善的。

System memory consolidation is a brain-wide process that transforms recently encodedliable memories into more stable and long-term memories. The 'two-stage' theory ofmemory process posits that memory consolidation engages bi-directional dialogueduring offline state between the hippocampus, where memory traces are initiallyformed, and the neocortex, where memory traces are eventually stored for long-termretention. During active exploration of a spatial context, sub-populations of neurons inboth CA1 region of hippocampus (HPC) and primary visual cortex (V1) develop spatialsensitivity with experience such that those spatially-modulated neurons would beactivated sequentially along animal's running trajectory. Strikingly, during offline period,the neuronal sequences observed in both regions during active spatial experience can bereplayed in a spatiotemporally coordinated fashion. Disruption of such replay inhippocampus would impair spatial memory performance. These observations led to thepossibility that interaction between the hippocampus and the neocortex around thetime of coherent replay may be instrumental to memory consolidation. However, as oftoday, we are unclear about the nature and direction of cortico-hippocampalinformation flow around the time of replay. Previous evidence suggested that change inneocortical activity can precede replay events. Nevertheless, it is unknown if the V1activity patterns prior to a replay event can selectively influence the memory content toreplay in both hippocampus and V1. Furthermore, we do not know whetherimprovement in coherence of hippocampal-cortical interaction during post-trainingreplay predict the memory performance and fidelity of the learned spatial informationin the V1.To address these research questions, we aim to record from V1 and CA1 of HPC whilemice learn to discriminate two novel linear tracks in virtual reality (VR). Two VR lineartracks with visually-identical landmarks in two positions will be used to trackdevelopment of spatial modulation in V1 over time. We will also train the animal to lickfor a reward when reaching a designated reward zone for each linear track, whichencourages animals to actively discriminate between two linear tracks. By recording V1and CA1 neural activities during multiple learning sessions and post-learning restsessions over the course of several weeks, we would like to examine how replaydynamics and cortico-hippocampal interactions during replay change as thediscriminatory memory performance improves over time.