Dynamically rich states in balanced networks induced by single-neuron dynamics

Abstract

Network states with rich dynamics and highly variable firing rates of individual neurons are prominent in experimental observations and thought to benefit complex information processing and learning. Such states have been proposed to arise from properties of network coupling, like a strong connectivity or slow synaptic dynamics. Here, we identify an alternative mechanism based on weak synaptic coupling and intrinsic cellular dynamics. We show that a switch in the cellular excitability class of action-potential generation (via a switch in the underlying mathematical bifurcation) results in super-Poissonian spiking variability in random balanced networks, where the recurrent network reinforces rather than averages out the single-cell effect. Information encoding is shifted to higher frequency bands and collective chaos in the network is enhanced when intrinsic cellular dynamics follow a saddle homoclinic orbit (HOM) bifurcation. Entropy production and attractor dimension peak close to the saddle-node-loop point that separates the two excitability classes, pointing to the qualitative nature of the transition as the origin of the effect. Because such switches are induced by common physiological variables, like temperature or extracellular potassium, and leave synaptic weights untouched, they offer a route to setting a network's dynamical regime independently of plasticity. The robust effect links the biophysics of individual neurons to collective dynamics of large random networks, highlighting the relevance of single-cell dynamics for computation in physiological and artificial networks.