Mechanisms of the phase locking of auditory-nerve fibers: a modeling approach

听觉系统非凡的时间精确度及其处理速度的一个标志是听神经纤维(ANF；初级传入)和一些更多的中枢神经元的神经元放电(棘波)对声刺激的精细结构的相位锁定。例如，在对音调的反应中，相位锁定反映在刺激周期内尖峰的分布，即使是那些200微米S或更短的尖峰(根据物种的限制)，时间离散低至30微米S。ANF的相位锁定对于声音定位和对音高的感知似乎是至关重要的，这是音乐、语音和其他发声的一个显著特征，也被用作推断耳蜗力学的工具。为此，我们开发了一个简约的现象学模型，该模型准确地解释了在刺激周期过程中CAT ANF的瞬时尖峰概率，作为声音频率和电平的函数。这个模型的一个关键要求是在前端有一个动态的Boltzmann传递函数，它可以快速适应主流的声级。我们还开发了一个突触-囊泡池耗尽和补充的模型，该模型准确地解释了这些ANF在自发活动期间的尖峰统计数据(尖峰间隔分布、序列间隔相关和尖峰计数分布)。我们现在计划将这两个模型结合起来，以也解释声音驱动的锁相响应的尖峰统计数据。我们还计划通过记录ANF对新的混合幅度刺激的响应并在逐个周期的基础上检查相位锁定来进一步表征Boltzmann传递函数的动力学。此外，我们计划通过将我们的模型应用于来自鸟类和具有条带突触改变的巴松管突变小鼠的数据来测试我们的模型的更普遍的有效性。我们的工作将有助于确定ANF反应显著的时间精确度背后的关键机制。

A hallmark of the extraordinary temporal precision of the auditory system and its processing speed is phase locking of neuronal discharges (spikes) of auditory-nerve fibers (ANFs; the primary afferents), and some more central neurons, to the fine-structure of acoustic stimuli. In response to tones, for example, phase locking is reflected in a peaked distribution of spikes over a stimulus cycle, even those that are 200 µs or shorter (the limit depending on species), with a temporal dispersion as low as 30 µs. Phase locking by ANFs appears crucial for sound localization and the perception of pitch, a salient feature of music, speech, and other vocalizations, and has also been used as a tool to infer cochlear mechanics. To this end, we developed a parsimonious phenomenological model that accurately accounts for the instantaneous spike probability of cat ANFs in the course of a stimulus cycle, as functions of sound frequency and level. A key requirement of this model is a dynamic Boltzmann transfer function at the front end, which rapidly adapts to the prevailing sound level. We also developed a model of synaptic-vesicle-pool depletion and replenishment that accurately accounts for spiking statistics (interspike interval distributions, serial interval correlations, and spike-count distributions) of these ANFs during spontaneous activity. We now plan to combine the two models to also account for the spiking statistics of sound-driven phase-locked responses. We also plan to further characterize the dynamics of the Boltzmann transfer function by recording ANF responses to novel mixed-amplitude stimuli and examining phase locking on a cycle-by-cycle basis. Furthermore, we plan to test the more general validity of our models by applying them to data from birds and from Bassoon mutant mice with altered ribbon synapses. Our work will help identify the key mechanisms underlying the remarkable temporal precision of ANF responses.