Failure in identification of overlapping spikes from multiple neuron activity causes artificial correlations

Area of Science:

• Neuroscience
• Computational Neuroscience
• Signal Processing

Background:

• Extracellular recordings often capture multiple neurons per electrode.
• Spike sorting methods aim to separate these neuronal signals.
• Overlapping spikes present a significant challenge for accurate signal separation.

Purpose of the Study:

• To define and analyze the 'shadowing effect' in multi-neuron recordings.
• To investigate artifacts introduced by spike overlap in auto- and cross-correlation functions.
• To improve the evaluation of neural firing patterns and synchronization from single-electrode recordings.

Main Methods:

• Analysis of extracellular recordings from the globus pallidus in behaving primates.
• Review of existing literature on spike sorting and neural synchrony.
• Simulations of independent model neurons to replicate observed effects.

Main Results:

• The 'shadowing effect' causes artifacts in auto- and cross-correlation, not just decreased zero-offset cross-correlation.
• These artifacts are pronounced in high firing rate or synchronous brain regions.
• Simulations confirm that observed artifacts arise from overlapping spikes, not necessarily neural interactions.

Conclusions:

• The 'shadowing effect' significantly impacts the interpretation of neural synchrony and firing patterns.
• Understanding and estimating these artifacts is crucial for accurate analysis of single-electrode multi-neuron recordings.
• This work provides a framework for better evaluating underlying neural activity despite spike overlap limitations.