单侧准备电位的统计分析方法

发布时间：2018-03-17 12:44

本文选题：脑电信号　切入点：单侧准备电位　出处：《哈尔滨师范大学》2015年硕士论文　论文类型：学位论文

【摘要】：认知神经科学是探究人脑神经机制的一门学科,已经成为现代科学研究的热门课题,时间相关电位(Event Related Potential, ERP)因其无创性和经济性而成为认知神经科学研究的主要手段.单侧准备电位是ERP的基本成分之一,它是指在随意运动中,反应效应器方位所对应的对侧大脑皮层出现的准备电位(Lateralized readiness potential, LRP)可以显示被试反应的正确率,还可以对运动预备的程度进行定量,即运动预备的程度高,则LRP波幅也高.LRP研究的重点是启动点和峰值的测量.LRP的启动点是划分反应阶段的重要的时程指标,LRP峰值是表征预备程度的指标.这两个指标的测量都是在获取LRP波形的基础上进行,而现实中,LRP波形的低信噪比大大影响了启动点和峰值的测量精度.传统的启动点测量方法有三类,基线漂移法,阈值法和分段回归法,基线漂移法和阈值法只用到LRP的局部信息进行推断,测量结果严重受到噪音的影响,相比之下,分段回归法是基于LRP的整体信息进行测量,但是,分段回归不能恰当的描述LRP的连续过程.本文提出了基于指数拟合的启动点测量方法,该方法能够克服噪音对测量的影响,又能体现LRP的连续过程.真实数据分析结果显示,本方法大大提高了测量精度.本文针对峰值提出基于滑动平均的测量方法.这一方法测量的结果更准确稳定.这两个方法的提出不仅对于不同试验条件下启动点差异的测量提供了更有效的途径,对单个试验条件下启动点的精准测量也是有效的.
[Abstract]:Cognitive neuroscience is a subject that explores the mechanism of human brain nerve, and has become a hot topic in modern scientific research. Event Related potential (ERP) has become the main method of cognitive neuroscience research because of its noninvasive and economical properties. Unilateral preparatory potential is one of the basic components of ERP, which refers to random exercise. The contralateral readiness potential (LRP) corresponding to the orientation of the response effector can show the correct rate of the reaction, and can also quantify the degree of motor preparation, that is, the degree of motor preparation is high. Then the research of LRP amplitude is also high. The emphasis of the research is that the starting point and the measurement of peak value. The starting point of LRP is an important time-history index to divide the reaction stage. The peak value of LRP is an indicator of the degree of preparation. The measurement of these two indexes is to obtain the LRP. Based on the waveform, In reality, the low signal-to-noise ratio (SNR) of LRP waveform greatly affects the measurement accuracy of starting point and peak value. There are three kinds of traditional starting point measurement methods: baseline drift method, threshold method and piecewise regression method. The baseline drift method and threshold method only use the local information of LRP to infer, and the measured results are seriously affected by noise. In contrast, the piecewise regression method is based on the whole information of LRP, but, Piecewise regression can not properly describe the continuous process of LRP. In this paper, a starting point measurement method based on exponential fitting is proposed, which can overcome the influence of noise on the measurement and reflect the continuous process of LRP. The results of real data analysis show that, This method has greatly improved the accuracy of measurement. In this paper, a new measuring method based on moving average for peak value is proposed. The results of this method are more accurate and stable. These two methods are not only for the starting point under different test conditions. The measurement of differences provides a more effective way, Accurate measurement of the starting point under a single test condition is also effective.
【学位授予单位】：哈尔滨师范大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：O213

【相似文献】