Balanced Oscillatory Coupling Improves Information Flow

Abstract

All animals are able to rapidly change their behavior. The neural basis of such flexibility requires that groups of distant neural ensembles rapidly alter communications with selectivity and fidelity. Low frequency oscillations are a strong candidate for how neurons coordinate communication via the dynamic instantiation of functional networks. These dynamic networks are argued to rapidly guide the flow of information, with the presumption that stronger oscillations more strongly influence information flow. Surprisingly, there is scant evidence or theoretical support for how oscillatory activity might enhance information flow. Here we introduce a novel computational model for oscillatory neural communication and show that, rather than the strength of the oscillation, it is the balance between excitatory and inhibitory neuronal activity that has the largest effect on information flow. When coupling between an oscillation and spiking has balanced excitatory-inhibitory inputs, information flow is enhanced via improved discriminability between signal and noise. In contrast, when coupling is unbalanced, driven either by excessive excitation or inhibition, information flow is obstructed, regardless of the strength of the oscillation. A multitude of neuropathologies, including Parkinsons disease, schizophrenia, and autism, are associated with oscillatory disruptions and excitation-inhibition imbalances. Our results show that understanding the distinction between balanced and unbalanced oscillatory coupling offers a unifying mechanistic framework for understanding effective neural communication and its disruption in neuropathology.