The coherence of neural activities among different areas in the brain has received great attention because it is valuable in understanding the functional mechanism of brain structures. While many methodologies, such as time-frequency and entropy analysis, have been applied to evaluate relations between neural signals, these techniques haven’t been effective in assessing neural communication in order to reach conclusions. Considering various measurements, the results analyzed by the above-mentioned algorithms may be influenced by the types of neural signals and their amplitudes, which affect their reliability and consistency. In this study, we introduced two new methods, phase-phase and spike train correlations, to analyze the neural signals communications among various areas of the brain, aiming to decipher neural information communications between different brain structures of normal rats and those with noise-induced tinnitus, a ringing condition in the ear or head. To test the proposed methodologies, a set of electrophysiological recordings of tinnitus-related spontaneous activities were conducted in the auditory cortex (AC), inferior colliculus (IC), and dorsal cochlear nucleus (DCN). The results using the two proposed algorithms were demonstrated and compared to those obtained by the transfer entropy (TE) method using the same experimental data set. Both algorithms yielded a result in a consistent scale of zero to one indicating the strength of correlation and showed a similar trend to results by TE. The experimental results on rats have shown information flow within and between most structures with a stronger correlation at lower frequencies.

中文翻译：

---

使用相位和尖峰训练相关算法评估听觉神经活动和信息传递。

大脑不同区域之间神经活动的连贯性受到了极大的关注，因为它对于理解大脑结构的功能机制很有价值。虽然许多方法，如时频和熵分析，已被用于评估神经信号之间的关系，但这些技术在评估神经通信以得出结论方面并不有效。考虑到各种测量，上述算法分析的结果可能会受到神经信号类型及其幅度的影响，从而影响其可靠性和一致性。在这项研究中，我们介绍了两种新方法，相位相位和脉冲序列相关性，来分析大脑各个区域之间的神经信号通信，旨在破译正常大鼠的不同大脑结构与噪音引起的耳鸣（耳朵或头部的响铃状况）之间的神经信息交流。为了测试所提出的方法，在听觉皮层 (AC)、下丘 (IC) 和背侧耳蜗核 (DCN) 中进行了一组耳鸣相关自发活动的电生理记录。使用相同的实验数据集演示了使用两种建议算法的结果，并与通过传递熵 (TE) 方法获得的结果进行了比较。两种算法都产生了一致的从零到一的结果，表明相关强度，并显示出与 TE 结果相似的趋势。