van Stokkum I H
Department of Medical Physics and Biophysics, University of Nijmegen, The Netherlands.
Hear Res. 1990 Jan;43(2-3):231-50. doi: 10.1016/0378-5955(90)90231-d.
In a previous paper [Van Stokkum and Gielen, Hear. Res. 41, 71-86, 1989] a model was presented to describe the processing of monaural stimuli by the auditory periphery of the grassfrog. The main components of this model were: a middle ear filter, transduction and tuning of the haircell, short-term adaptation, action potential (event) generation with refractory properties, and spatiotemporal integration of converging inputs. The model is now extended to model auditory midbrain neurons as third order neurons. The mechanisms that generate selectivity for temporal characteristics of sound are adaptation, coincidence detection of second order neurons, temporal integration of third order neurons, and most important, event generation of the first, second and third order model neurons. Variation of the parameters of the model successfully reproduces the range of response patterns which have been obtained from eighth nerve fibres, dorsal medullary nucleus neurons, and torus semicircularis neurons without inhibition. With a single set of parameters the output of the model in response to a set of spectrally and temporally structured stimuli qualitatively resembles the responses of a single neuron to all these stimuli. In this way the responses to the different stimuli are synthesized into a framework, which functionally describes the neuron.
在之前的一篇论文中[范·斯托库姆和吉伦，《听觉研究》41卷，71 - 86页，1989年]，提出了一个模型来描述草蛙听觉外周对单耳刺激的处理过程。该模型的主要组成部分包括：中耳滤波器、毛细胞的转导与调谐、短期适应、具有不应期特性的动作电位（事件）产生以及汇聚输入的时空整合。现在该模型扩展为将听觉中脑神经元模拟为三阶神经元。产生对声音时间特征选择性的机制包括适应、二阶神经元的重合检测、三阶神经元的时间整合，以及最重要的是一阶、二阶和三阶模型神经元的事件产生。模型参数的变化成功再现了从无抑制的第八神经纤维、背侧髓核神经元和半规管神经元获得的一系列反应模式。使用一组参数，模型对一组频谱和时间结构刺激的响应在质量上类似于单个神经元对所有这些刺激的响应。通过这种方式，对不同刺激的响应被综合到一个框架中，该框架从功能上描述了神经元。
