Flexibility vs consistency: Quantifying differences in neuromodulatory elicited patterns of activity

Central pattern generating circuits underly fundamental behaviors such as respiration or locomotion and are under the influence of neuromodulators. The presence of neuromodulators is thought to confer flexibility to these circuits to generate distinct patterns of activity to meet distinct behavioral needs. Network output flexibility can be achieved by distinct classes of neuromodulators, those which have convergent cellular actions but divergent circuit actions or by those which have divergent cellular actions but convergent circuit actions. Both classes of neuromodulator exist in the stomatogastric nervous system of the crab Cancer borealis and influence the activity of a central pattern generating circuit in the stomatogastric ganglion, the pyloric network. The ability of both classes of neuromodulator, when applied individually, to generate qualitatively and quantitatively distinct patterns of activity has been demonstrated with respect to a baseline activity state. While it is assumed that each individual neuromodulator’s activity pattern is distinct, there has yet to be a fully quantitative description of the degree of difference between two modulated activity patterns. It is also unlikely that any single circuit will be under the influence of only a single neuromodulator at any point. Therefore, the possibility of generating distinct network outputs increases with each distinct combination of neuromodulators. While the actions of individual neuromodulators have been explored, the consequences of co-modulation on the pyloric network’s output are less understood. Previous attempts at quantifying the effects of a neuromodulator on the pyloric network output relied on evaluating only a single, often multi-dimensional, attribute of activity at a time and statistically testing the dependent parameters of that attribute with statistics that assume independence. This dissertation uses a new approach to quantify and statistically test how different one neuromodulator elicited pattern of activity is from another, preserving the inherent multi-dimensional nature of the attributes evaluated. The results of this dissertation show that the pyloric network output is able to generate statistically distinct network outputs with individual neuromodulators; however, flexibility is lost in favor of consistency under co-modulatory conditions