Document Actions

You are here: Home / The Network / NNCN / Bernstein Collaborations / Olfactory Coding

Olfactory Coding

Bernstein Collaboration for Computational Neuroscience (BCOL)

Subproject 1: Primary olfactory coding: the honeybee antennal lobe

Olfactory coding has to accomplish a formidable task in animals: on one hand, odors need to be identified, in order to recognize odor sources reliably. On the other hand, the identity of an odor needs to be judged with sufficient flexibility to account for natural variations of odors. The theoretical capacity limits of the olfactory system will be evaluated quantitatively. To this end, theoretical and experimental approaches will be combined in order to understand the mechanisms of olfactory coding in animal brains. In the experimental approach, odor-evoked activity patterns will be recorded in the brain of honeybees using optical imaging methods with high spatial and temporal resolution. It is expected that the data will provide an opportunity in order to characterize a complex olfactory system across all information-coding channels (the glomeruli). By extrapolation from these models, and by analyzing the capacity limits, new insight into the computational principles governing brain function will be gained.

Subproject 2: Higher olfactory processing: the honeybee mushroom bodies

The insect olfactory system is an ideal model to understand natural neural networks in their entirety, from sensory coding to behavioural control. The modular structure of the processing steps reaches from the sensory neurons, via glomeruli in the antennal lobes, to the mushroom body microcircuits. This architecture is particularly interesting for theoretical models that seek to create realistic simulations of brain activity. In the experimental part single neurons in the antennal lobe and in the mushroom bodies will be characterized using electrophysiological and optophysiological recordings, and the morphological branching will be mapped with confocal microscopy. Models of the olfactory system will be created on this basis. These detailed models will help to follow olfactory processing of odor identity and odor concentration along its processing steps, and to understand changes caused by learning.