Tononi G, Sporns O, Edelman G M
The Neurosciences Institute, San Diego, CA 92121, USA.
Proc Natl Acad Sci U S A. 1996 Apr 16;93(8):3422-7. doi: 10.1073/pnas.93.8.3422.
We have previously derived a theoretical measure of neural complexity (CN) in an attempt to characterize functional connectivity in the brain. CN measures the amount and heterogeneity of statistical correlations within a neural system in terms of the mutual information between subsets of its units. CN was initially used to characterize the functional connectivity of a neural system isolated from the environment. In the present paper, we introduce a related statistical measure, matching complexity (CM), which reflects the change in CN that occurs after a neural system receives signals from the environment. CM measures how well the ensemble of intrinsic correlations within a neural system fits the statistical structure of the sensory input. We show that CM is low when the intrinsic connectivity of a simulated cortical area is randomly organized. Conversely, CM is high when the intrinsic connectivity is modified so as to differentially amplify those intrinsic correlations that happen to be enhanced by sensory input. When the input is represented by an individual stimulus, a positive value of CM indicates that the limited mutual information between sensory sheets sampling the stimulus and the rest of the brain triggers a large increase in the mutual information between many functionally specialized subsets within the brain. In this way, a complex brain can deal with context and go "beyond the information given."
我们之前推导了一种神经复杂性（CN）的理论度量方法，旨在刻画大脑中的功能连接。CN从神经系统内各单元子集之间的互信息角度，测量神经系统内统计相关性的数量和异质性。CN最初用于刻画与环境隔离的神经系统的功能连接。在本文中，我们引入一种相关的统计度量方法，匹配复杂性（CM），它反映了神经系统从环境接收信号后CN所发生的变化。CM测量神经系统内固有相关性的集合与感觉输入的统计结构的拟合程度。我们表明，当模拟皮层区域的固有连接随机组织时，CM较低。相反，当固有连接被修改以差异性放大那些恰好被感觉输入增强的固有相关性时，CM较高。当输入由单个刺激表示时，CM的正值表明采样该刺激的感觉层与大脑其余部分之间有限的互信息会引发大脑内许多功能特化子集之间的互信息大幅增加。通过这种方式，复杂的大脑能够处理上下文并“超越所给信息”。
