Hard to see, yet critical to function: how neuronal excitability classes add to neuronal diversity

Neuronal diversity arises from cellular properties that differ between neurons. Not all such differences, however, are obvious at first glance. In this talk, I will focus on the mechanism of action-potential generation and the distinct excitability classes found in regularly firing neurons. While action potentials of distinct classes are nearly indistinguishable to the naked eye, in conductance-based models as well as biological neurons they arise from three distinct mathematical bifurcations. The specific bifurcation underlying spike generation qualitatively influences how information is processed and which dynamical states can emerge in neural networks. Interestingly, many physiological parameters, including temperature and ionic concentrations, can shift a neuron's excitability class, often leading to significant changes in circuit function. I will present examples of the functional implications of different excitability classes in Drosophila and rodents, explore how transitions between them contribute to computational flexibility and robustness, and predict how neurons might stabilize their excitability class, akin to homeostatic regulation of firing rate.