A computational model for signal processing by the dorsal cochlear nucleus. I. Responses to pure tones.

Much information is available on the anatomical organization and neurophysiological properties of the major cell types in the dorsal cochlear nucleus (DCN). The complicated response properties of individual cells and units in the DCN indicate that substantial information processing already occurs at the level of the DCN. A large number of connectional hypotheses have been put forward to explain various aspects of the response characteristics of DCN cells, but many of the consequences of these hypotheses have not been investigated quantitatively. In this paper, we investigate these hypotheses by constructing and testing mathematical and computational models and compare our results to those of previous modeling studies. The simplest versions of our models include auditory nerve (AN) fibers, type II cells (inhibitory interneurons) and type IV cells (fusiform and giant cells). The model response maps, i.e., the pattern of output of model type IV cells, generated by the simplest model have some but not all the features of the experimental response maps of type IV neurons. In particular, the excitatory region which occurs at best frequency is not isolated and the excitatory region at low frequencies and high amplitude is narrower than observed. Since experimental evidence exists that some of the connections between these cell types are divergent or convergent across adjacent isofrequency sheets, the effect of such convergence and divergence was then investigated. Response maps so obtained reproduce many of the qualitative features of the experimental maps. Enlargement of the model by including inhibitory interneurons (e.g., stellate cells) that receive convergent input from AN fibers and which inhibit type II cells results in the generation of response maps that, with some variations in connectional patterns and physiological properties of neurons, match most of the essential features seen in the large variety of experimental maps obtained from the cat DCN.

Duke Scholars

Author Michael C. Reed Mathematics