Sensory sequence representation and discrimination in cortical circuits

我们在空间和时间上体验世界。为了理解环境，我们需要能够区分序列，即时间模式的刺激。听觉和触觉刺激只能通过它们的时间结构来理解--这表明大脑特别擅长破译和记忆，例如当我们学习一首新歌或某种节奏时。产生和解码电活动序列的能力在整个大脑中无处不在。此外，在包括抑郁症和精神分裂症在内的精神障碍中，序列学习和时间持续时间的处理受到损害。然而，尽管序列选择性具有这种中心性，但人们对其在大脑中的底物知之甚少。从一个连续的时间模式第一次反映在神经元的电活动中，到它被归类为“对象”，一个有意义的实体，这一连串的事件仍然未知。这是我们对大脑功能理解上的一个重大差距。这个项目的目标是了解大脑皮层神经元群体的活动如何产生序列的表示。大脑皮层是大脑的一部分，它结合来自感觉的信息，赋予它一个含义，将它与早期的经验联系起来，并决定一个反应。为了显示神经元中的电活动如何与序列识别有关，有必要将序列呈现过程中的神经元反应与特定行为联系起来，以表明受试者区分序列。为此，我们创造了一种新颖的行为任务，在老鼠和人类身上都成功地完成了这项任务。老鼠学会察觉胡须--触觉的“歌”或“词”--传递给它们的任意刺激序列。呈现的序列只在时间模式上有所不同，小鼠学习识别特定的序列以获得奖励。这种实验设计的目的是让我们能够在大脑中搜索序列识别的一般原理，特别是测试对给定序列敏感的神经元是否驻留在大脑皮层的感觉区域。我们建议将这一设计与强大的新技术相结合，以检查和操纵大脑活动。我们将使用我们实验室已经建立的电生理学和双光子成像方法来记录活动。录音将通过最先进的遗传和成像工具针对大脑皮层的神经元组。这些实验将使我们能够发现与受控感觉行为相关的神经元活动，并识别对特定序列具有选择性的神经元。我们的研究将为神经元电路如何随着时间的推移处理信息以产生感觉和知觉提供新的见解。由于大脑皮层感觉区域的神经元回路在不同的哺乳动物物种中是相似的，这一新知识将在不同物种之间相关，也将适用于人类序列学习。这一结果的意义不仅仅在于改善我们对大脑皮层神经元信息流的基本理解。由于序列处理经常在扰乱大脑皮质神经元电路正常建立的障碍中受损，我们的方法将有助于理解这些障碍的功能影响，并为测试大脑病理的神经生物学模型提供一种分析方法。

We experience the world in both space and time. To make sense of an environment, we need to be able to distinguish sequences, i.e., stimuli that are temporally patterned. Auditory and tactile stimuli can only be understood in terms of their temporal structure - a cue that the brain is particularly good at deciphering and remembering, for example when we learn a new song or a certain rhythm. The ability to generate and decode sequences of electrical activity is ubiquitous throughout the brain. Moreover, sequence learning and the processing of temporal duration are impaired in psychiatric disorders including depression and schizophrenia. Yet despite this centrality of sequence selectivity, remarkably little is known about its substrates in the brain. The chain of events from when a sequential temporal pattern is first reflected in the electrical activity of neurons to when it is classified as an "object," a meaningful entity, remains unknown. This is a major gap in our functional understanding of the brain. The goal of this project is to understand how representations of sequences emerge from the activity of populations of neurons in the cerebral cortex - the part of the brain that combines information from the senses, assigns it a meaning, relates it to earlier experience and decides on a response. To show how electrical activity in neurons relates to sequence recognition, it is necessary to link neuronal responses during presentation of a sequence to a specific behaviour showing that the subject distinguishes the sequence. To this end, we have created a novel behavioural task that is successfully performed by mice as well as by humans. Mice learn to detect arbitrary sequences of stimulation delivered to their whisker - tactile "songs" or "words." The sequences presented differ only in their temporal patterning, and mice learn to recognise a particular sequence for a reward. This experimental design is designed to allow us to search for general principles of sequence recognition in the brain, and specifically to test whether neurons sensitive to a given sequence reside in sensory areas of the cortex. We propose to combine this design with powerful new techniques for examining and manipulating brain activity. We will use methods of electrophysiology and two-photon imaging, already established in our laboratory, to record activity. Recordings will be targeted to groups of neurons in the cerebral cortex by means of state-of-the-art genetic and imaging tools. These experiments will allow us to uncover the neuronal activity related to the controlled sensory behaviour and identify neurons that are selective to a particular sequence. Our research will provide new insights into how neuronal circuits process information over time to give rise to sensation and perception. Because the neuronal circuitry in sensory areas of the cerebral cortex is similar in different mammalian species, this new knowledge will be relevant across species and will apply to human sequence learning as well. The outcome will have implications beyond improving our fundamental understanding of information flow through cortical neurons. As sequence processing is often impaired in disorders that perturb the normal establishment of neuronal circuits in the cortex, our approach will help the understanding of the functional effects of those disorders, and provide an assay for testing neurobiological models of brain pathologies.