[1]师黎,许昆峰,牛晓可.基于状态空间模型的神经元动态相关性研究[J].郑州大学学报(工学版),2015,36(01):1-5.[doi:10.3969/ j.issn. 1671 -6833.2015.01.001]
　SHI Li,XU Kun-feng,NIU Xiao-ke.The analysis of dynamic correlation between neurons based on state-space log-linear model[J].Journal of Zhengzhou University (Engineering Science),2015,36(01):1-5.[doi:10.3969/ j.issn. 1671 -6833.2015.01.001]
点击复制

基于状态空间模型的神经元动态相关性研究()
分享到：

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
36
期数:
2015年01期
页码:
1-5
栏目:
出版日期:
2015-01-10

文章信息/Info

Title:
The analysis of dynamic correlation between neurons based on state-space log-linear model
作者:
师黎; 许昆峰; 牛晓可
郑州大学电气工程学院,河南郑州450001
Author(s):
SHI Li; XU Kun-feng; NIU Xiao-ke
School of Electrical Engineering,Zhengzhou University,Zhengzhou 450001,China
关键词:
状态空间对数线性模型; 动态相关性; 信息编码; 同步作用
Keywords:
state-space log-linear model; dynamic correlation; information coding;  synchronization
分类号:
TN911.7
DOI:
10.3969/ j.issn. 1671 -6833.2015.01.001
文献标志码:
A
摘要:
神经元间相关性的研究是深入理解神经元集群信息传递与编码机理的基础.首先,采用状态空间对数线性模型初步估计神经元间的动态相关性,针对输入数据特征对模型估计值置信区间的影响,提出了通过筛选数据优化置信区间来提高模型估计精度.然后,通过提取动态相关性的特征,分析神经元间相关性在不同朝向光栅刺激下的动态特性,进而研究了神经元间同步作用对视觉刺激信息的编码作用.最后,在麻醉的Long Evens( LE)大鼠初级视觉皮层上进行了实验验证.结果表明:采用剔除发放率偏小的序列的数据筛选方案能够有效地提高模型估计值的精度;神经元间的锋电位同步作用对朝向光栅刺激信息具有一定的编码作用.
Abstract:
The research on correlation between neurons is the foundation to understand the mechanism of infor-mation transmission and coding of neuronal population. A novel method called state-space log-linear model wasused to estimate the dynamic correlation between paired neurons,and data sieving methods were proposed toimprove the accuracy of model results for the effects of input data characteristics on the confidence interval ofthe model estimated values. By extracting the characteristics of dynamic correlation curves,changing charac-teristics of paired neurons’correlation was analyzed and then the effect on information coding of visual stimu-lus from synchronization between paired neurons was studied. Experimental verification was carried out in theprimary visual cortex of anesthetized rats.The results show that: the accuracy of the estimated value of themodel can be improved by removing the data with small firing rates,and synchronization between paired neu-rons encodes the information of different grating stimuli.
更新日期/Last Update: