Time-varying phase relationship between spiking neuronal responses and local field potentials in olfactory processing in manduca sexta

The Transient Oscillatory Model (TOM) seeks to explain neural coding in the first processing center of the invertebrate olfactory system - the antennal lobe - by proposing that activated primary output neurons participate in the encoding ensemble only transiently, based on the coincidence of their spiking relative to local field potential (LFP) oscillation phase. Using the sphinx moth, Manduca sexta, as a model system, we tested the predictions of the TOM with regard to the trial-to-trial consistency of odor driven oscillations, the phase-locking of action potentials to LFP oscillations, and the role of GABAA receptors in the generation of these oscillations. We quantified the changes in LFP oscillation frequency over time via time-frequency representation (TFR) analysis, and calculated the timing of action potentials relative to the phase of the LFP oscillations (vector-strength analysis). In accordance to the TOM\u27s predictions for the role of LFP oscillations, TFRs revealed that odor-driven oscillations modulate in a stimulus specific manner. However, contrasting with these same predictions, vector-strength analyses illustrated that phase locking was higher during spontaneous activity than during odor response. Finally, in disagreement with TOM predictions, disruption of GABA A receptors by BMI application reduced odor-driven LFP oscillations across time-periods and regions of the AL. Further BMI application reduced phase-locking of action potentials to the LFP, while leaving phase-locking highest during spontaneous activity. Consequently, we find the overall architecture of the TOM to be incompatible with these findings