A simulation system and model for the anuran retina.

详细信息   

• 作者：Teeters ; Jeffrey Lee.
• 学历：Doctor
• 年：1989
• 导师：Arbib, Michael A.
• 毕业院校：University of Massachusetts
• 专业：Computer Science.;Biology, Neuroscience.
• CBH：9001573
• Country：USA
• 语种：English
• FileSize：15681420
• Pages：278

文摘

There are two parts to the thesis: a simulation system and a model of the anuran retina. The simulation system, called MDL for Model Development Language, uses an assignment statement notation tailored for 2-dimensional layered neural networks. The key features are: (1) Operations occur on 2-dimensional arrays which correspond to the layout of many brain structures. (2) A spatial convolution operation models spatial summation occurring in dendritic trees. (3) Time invariant expressions and differential equations are specified in a manner that removes the numerical methods from the model. These features allow high level modeling using leaky integrator model neurons and modeling of detailed biophysical properties, such as the Hodgkin Huxley equations and voltage clamp experiments. Conciseness of the language allows specification of a moderately complex retina model in only one or two pages.;The retina model consists of models for each of the five major anuran ganglion cell types. A common model OPL (Outer Plexiform Layer) containing receptors, horizontals, and bipolar cells generates input to the model Inner Plexiform layer (IPL). Processing in the IPL specific to each ganglion cell type forms the model which qualitatively accounts for many characteristic ganglion cell response properties. The class 2, 3, and 4 cell models also quantitatively account for response dependence on extent and orientation of moving rectangular stimuli and dependence on stimulus velocity.;The model shows that horizontal cells are not crucial for most receptive field properties, but may control sensitivity of bipolar cells in a way that prevents class 1 and class 2 cells from responding to full field illumination changes. For class 1, 3, and 4 cells, a simple weighted summation of bipolar cell and differentiated bipolar cell outputs is sufficient to explain responses, although in the class 3 cell, the onset of inhibition must be delayed with respect to excitation. The class 2 cell model requires a more complex means of summing IPL inputs to account for erasability and neuronal adaptation. Differences in transient amacrine cells feeding to each ganglion cell type may underlie variation in velocity dependence. The model yields suggestions for future experiments and modeling.