DYNAMICS OF NEURONAL FIRING CORRELATION - MODULATION OF EFFECTIVE CONNECTIVITY
DYNAMICS OF NEURONAL FIRING CORRELATION - MODULATION OF EFFECTIVE CONNECTIVITY
复制标题
DOI：
10.1152/jn.1989.61.5.900
复制
发表时间：
1989-05-01
影响因子：
2.5
通讯作者：
PALM, G
中科院分区：
文献类型：
--
作者：
AERTSEN, AMHJ;GERSTEIN, GL;PALM, G
1.我们重新审视的可能性，分析和解释的时间过程中的相关性，同时和单独记录两个神经元的尖峰列车。2.我们开发的程序来量化和适当规范的经典关节刺激周围时间散点图。这些允许将“原始”相关性分离成由单神经元放电率的直接刺激调制引起的分量和由两个神经元之间的各种类型的相互作用引起的分量。3.一个新开发的显着性测试（“惊喜”），以评估这样的推论。4.应用新的程序模拟穗列车允许已知的电路恢复。特别是，它被证明可以恢复快速刺激锁定调制的“有效连接”，即使他们被掩盖了强大的直接刺激调制的个人放电率。因此，这些程序提出了一个明显优于常用的“移位预测”的上级替代方案。“5.采用基于模型的方法，我们概括的经典措施，量化直接的神经元间的连接（“功效”和“贡献”），包括可能的刺激锁定的时间变化。6.应用新的程序，真实的穗列车从几个不同的准备表明，快速刺激锁定调制的“有效连接”也发生真实的神经元。
1. We reexamine the possibilities for analyzing and interpreting the time course of correlation in spike trains simultaneously and separably recorded from two neurons. 2. We develop procedures to quantify and properly normalize the classical joint peristimulus time scatter diagram. These allow separation of the "raw" correlation into components caused by direct stimulus modulations of the single-neuron firing rates and those caused by various types of interaction between the two neurons. 3. A newly developed significance test ("surprise") is applied to evaluate such inferences. 4. Application of the new procedures to simulated spike trains allowed the recovery of the known circuitry. In particular, it proved possible to recover fast stimulus-locked modulations of "effective connectivity," even if they were masked by strong direct stimulus modulations of individual firing rates. These procedures thus present a clearly superior alternative to the commonly used "shift predictor." 5. Adopting a model-based approach, we generalize the classical measures for quantifying a direct interneuronal connection ("efficacy" and "contribution") to include possible stimulus-locked time variations. 6. Application of the new procedures to real spike trains from several different preparations showed that fast stimulus-locked modulations of "effective connectivity" also occur for real neurons.