Studying spike trains using a van Rossum metric with a synapse-like filter

Area of Science:

• Computational Neuroscience
• Systems Neuroscience
• Neurophysiology

Background:

• Spike trains, the primary mode of neural communication, exhibit variability even for identical stimuli.
• Assessing the information content of individual spikes within these variable spike trains is a significant challenge.
• Existing spike train metrics have been used to determine the temporal precision relevant for information encoding.

Purpose of the Study:

• To develop a novel spike train metric that incorporates synaptic dynamics, specifically binding site depletion.
• To evaluate the effectiveness of this new metric in classifying neuronal responses based on stimuli.
• To investigate whether spike significance in encoding information is modulated by the proximity of preceding spikes.

Main Methods:

• A new spike train metric was constructed using a biophysical model of synaptic conductances.
• The metric explicitly includes the phenomenon of binding site depletion, where recent activity affects synaptic efficacy.
• The metric was applied to electrophysiological recordings from the primary auditory cortex of zebra finches.

Main Results:

• The developed metric effectively classified neuronal responses according to different stimuli in the dataset.
• The metric demonstrated that the impact of an individual spike on the calculated distance is reduced if it follows other spikes closely.
• This suggests that the significance of a spike in encoding information is modulated by its temporal relationship with prior spikes.

Conclusions:

• The new spike train metric, incorporating binding site depletion, is effective for analyzing neural data.
• Spike train properties, specifically the temporal context of spikes, play a crucial role in information coding.
• The proximity of spikes modulates their significance, representing a potential information-coding strategy in neural systems.