The olfactory bulb sits at the crossroads of input from an animal’s external and internal world. In this neural structure, chemical information from the environment interacts with contextual information emanating from higher cortical regions to shape mental representations of odor. Nevertheless, the factors influencing this interaction, and how the cortex manipulates these factors to the advantage of the animal, remain a mystery. To investigate this question, we have developed a large-scale computational model of the olfactory bulb. This model consists of a new algorithm to determine connectivity between mitral cells and granule cells, based in known anatomical constraints, combined with a dynamical systems approach utilizing the Izhikevich equations to simulate the network’s behavior. Using this model, we first examine connectivity and activity patterns of our network to demonstrate the strong relationship between structure and function in the olfactory bulb. We then further employ this model to analyze the effects of centrifugal feedback to the olfactory bulb on cortical odor representations; through this analysis, we are able to show that stochastic feedback patterns can evoke distinct trends in convergence and divergence between these representations depending on cortical excitability. Finally, we take advantage of the ease of incorporating new neurons into the model to study neurogenesis in the olfactory bulb, in particular to elucidate possible rules governing the placement of new cells. Through these experiments, our model provides new insight into the olfactory bulb and its role in the greater olfactory system.