ENCODING OF OLFACTORY INFORMATION WITH OSCILLATING NEURAL ASSEMBLIES

In the brain, fast oscillations of local field potentials, which are t hought to arise from the coherent and rhythmic activity of large numbe rs of neurons, were observed first in the olfactory system and have si nce been described in many neocortical areas. The importance of these oscillations in information coding, however, is controversial. Here, l ocal field potential and intracellular recordings were obtained from t he antennal lobe and mushroom body of the locust Schistocerca american a. Different odors evoked coherent oscillations in different, but usua lly overlapping, ensembles of neurons. The phase of firing of individu al neurons relative to the population was not dependent on the odor. T he components of a coherently oscillating ensemble of neurons changed over the duration of a single exposure to an odor. It is thus proposed that odors are encoded by specific but dynamic assemblies of coherent ly oscillating neurons. Such distributed and temporal representation o f complex sensory signals may facilitate combinatorial coding and asso ciative learning in these, and possibly other, sensory networks.