Convolution Metric for Neuron Membrane Potential Recordings

TL;DR

This paper introduces a convolution-based metric for comparing neuron membrane potential recordings, capturing subthreshold features, spike timing, and spike count in a single measure, improving upon previous spike train metrics.

Contribution

It proposes a novel convolution metric that incorporates membrane potential features and spike timing, with a carefully designed kernel for better sensitivity to small differences.

Findings

The metric effectively captures differences in membrane potential and spike timing.

It is first order in differences of spike times and membrane potentials.

The kernel design enhances frequency response and sensitivity.

Abstract

I provide a convolution metric which takes neural membrane potential recordings as arguments and compares their subthreshold features along with the timing and number of spikes within them--summarizing differences in these with a single "distance" between the recordings. Based on van Rossum's 2001 metric for spike trains, the metric relies on a convolution operation that it performs on the input data. The kernel used for the convolution is carefully chosen such that it produces a desirable frequency space response and, unlike van Rossum's kernel, causes the metric to be first order both in differences between nearby spike times and in differences between same-time membrane potential values: an important trait. 31 pages, 4 figures.