Biological neural networks: an application to odor information processing

Abstract

Traditionally, there has been a lot of research on how the brain manages visual and auditory inputs, but not that much attention has been devoted to the olfactory sensory system due, in part, to the complex nature of volatile organic compounds (VOCs). The aim of this project is to understand and build a computational neural model of the first stage of the olfactory pathway: the glomerular layer. Some experts in the field, supported by experimental evidence, point at this first stage as the neural location where an important aspect of odor information processing takes place: odor segregation. Our working hypothesis states that this input layer performs normalization and contrast and concentration enhancement over odorants before the odorant information is conveyed to deeper olfactory structures for detection. By performing an experiment on the computed neural network, we will be able to determine whether and how this odor segregation occurs. This is achieved by statistical analysis of the structure of the high-dimensional output space, understood as the spiking frequency and synchronization of glomerular neurons firing patterns.