以果蝇为模型探讨非傅立叶运动感知神经机制

Fourier motion is visual stimulus that consists of elements continuously moving in a direction. Both real motion of objects in the world and apparent motion in cinema belong to Fourier stimuli, which can be detected by a correlation-type movement detector. In contrast, the stimuli without continuously moving elements are called non-Fourier motion or higher-order motion, like those motion signals carried by variations in second-order image characteristics such as contrast, flicker, and texture. An important question is why higher-order stimuli are visible to a wide variety of animal species including humans, zebrafish and even Drosophila, given that such motion lacks coherent spatiotemporal correlations in luminance and therefore cannot be detected by a correlation-type movement detector. The question is essential and critical for revealing the mechanisms of motion processing, however, it remains open to date. Fruit fly receives emotion signals as major visual input and possesses a relatively simple neural system. In this project we investigate the neural mechanisms underlying non-Fourier motion perception by taking advantage of the rich genetic tools, quantitative behavioral analysis paradigm, and electrophysiological recoding techniques already available in Drosophila. By combining a computational modeling approach with experimental approach like behavioral experiments, the following topics will be addressed. Whether and how are the deep neuropils of optic lobe and the central brain involved in non-Fourier motion processing? Are there two distinct motion processing pathways, responsible for Fourier and non-Fourier motion detection, respectively? Whether or how does non-Fourier motion processing depend on object perception? The results will be expected to shed light on the general mechanisms underlying motion perception.

包含朝一个方向连续移动元素的视觉刺激叫傅里叶运动。真实物体运动及电影似动都属于这一类，可被相关型初级运动检测器检测到。与此相反，不包含连续移动元素的刺激称为非傅里叶或高阶运动，例如由对比度等局部特征变化定义的运动。一个有趣而重要的问题是：既然高阶运动因不含任何亮度变化而不能被运动检测器测到，为何人类甚至斑马鱼和果蝇都可以非常清楚地感受到这种运动？这类问题涉及到运动感知的根本机制，目前远未清楚。考虑到果蝇主要视觉来源为运动视觉，具有简约的神经系统、丰富的基因操作工具和方便的定量行为分析范式，且活体电生理技术已成功用于果蝇，本项目选择果蝇作为模型，采用神经环路建摸、计算仿真、行为实验等相结合的途径，深入探讨视叶深层和中央脑是否及如何参与非傅里叶运动感知，非傅里叶运动与傅里叶运动处理通路是否及如何互作用，运动感知与物体识别神经机制的内在关系等。研究结果期待有助于揭示运动信息编码和提取的普遍机制。

非傅里叶运动或称高阶运动没有亮度的时空变化，因而不能被相关型初级运动检测器测到。有趣的是，不仅人类可以非常清楚地感受到高阶运动，而且没有视皮层的斑马鱼和果蝇也可以有效感知这种高阶运动。目前对于主要视觉来源是运动视觉的各种蝇类的高阶运动感知机制知之甚少。本项目以果蝇为模式动物，采用神经元建摸、计算模拟、行为实验、基因操作等手段相结合途径，探讨非傅里叶运动信息处理的神经机制。我们在解析果蝇高阶运动神经回路机制、神经元树突计算机制等方面取得了进展。虽然“揭示果蝇高阶运动处理的神经机制”作为本项目的终极研究目标还没有完成，但是我们已经开展了坚实的前期探索工作，几个关键问题的研究目前正在深入进行中。研究结果有助于在神经元、神经回路和行为等水平理解果蝇运动信息编码和提取的普遍机制。..在项目执行过程中也进行了一些没有列入原计划的研究。例如我们在多巴胺投射区-纹状体Beta和Gamma节律产生机制方面做了深入探讨。纹状体90%以上的神经元是GABA能神经元，由主要表达多巴胺D1受体和主要表达多巴胺D2受体的两种中型多棘投射神经元和少量快速发放的中间神经元构成。一个重要的问题是，Beta和Gamma节律是内生的还是源于输入？我们构造了结构与参数与当前电生理测量吻合的大型纹状体神经网络，采用理论分析和大规模数值模拟方法，发现由非同步发放的三种GABA能神经元组成的纹状体可以内生涌现出特定频率振荡。与同领域研究比较，我们在纹状体振荡与快速发放GABA能中间神经元的依赖关系、去级化GABA电流功能、异质GABA能神经网络振荡机制等方面取得了进展。