Theory of optimal balance between excitation and inhibition can predict properties of synaptic inhibition

Abstract

Neuron’s synaptic input causes spike output depending on membrane excitability. In understanding the meaning of spikes, conventional approach is statistical analysis entirely depending on an input-output analysis without consideration of events happening within the neuron. In my thesis, I interpret the meaning of spikes with considering neuron as an observer who knows only internal biophysical state. According to theory proposed in my thesis, membrane excitability is an expectation of excitatory postsynaptic conductance (EPSG) amplitude, and a spike is generated only when EPSG amplitude exceeds its expectation (“prediction error”). When the membrane excitability is optimal by the accurate expectation about EPSG amplitude, optimal balance would be achieved by the peak of an excitatory postsynaptic potential (EPSP) being at precisely spike threshold, so that the spike generation is maximally sensitive to EPSG amplitude. It will maximize information about the EPSG amplitude in the binary spike output. Optimal balance would be implemented by a diversity of synaptic inputs and voltage-gated ion channels. In my thesis, I focused on optimal balance between synaptic excitation and inhibition. Synaptic inhibition counterbalance synaptic excitation, but it is not known what constitute optimal balance. I predicted inhibitory postsynaptic conductance (IPSG) that achieve optimal balance, using the theory (EPSP peak should be at spike threshold) and computer simulation. A single combination of IPSG amplitude and decay time was identified for a particular EPSG pattern that was distinct with respect to mean rate. The optimal IPSG increases in amplitude and decay rate as the EPSG rate increased from 1 to 800 Hz. As further proposed by the theory, the optimal IPSG parameters could be learned through anti-Hebbian rules. These theoretical predictions of IPSG parameters match experimental observations. From an infinite range of possible decay times, the theory predicted experimental data within less than a factor of 2. Thus the theory can explain a biophysical quantity from first principles.