Correlation between neural spike trains increases with firing rate.

Area of Science:

• Neuroscience
• Computational Neuroscience

Background:

• Temporal correlation of neuronal action potentials (spike trains) is observed across various brain regions.
• Correlated firing is implicated in crucial functions like stimulus encoding, attention, and motor control.
• The precise mechanisms and coding implications of correlated spiking remain incompletely understood, particularly the influence of input characteristics.

Purpose of the Study:

• To investigate whether spike train output correlation depends solely on input current correlation or also on input mean and variance.
• To elucidate the relationship between input correlation, firing rate, and output correlation in neuronal networks.

Main Methods:

• Computed spike train correlation coefficient for unconnected pairs of in vitro cortical neurons receiving correlated inputs.
• Employed analytical techniques and numerical simulations using 'integrate-and-fire' neuron models.
• Validated findings using a standard threshold-linear neuron model.

Main Results:

• Spike train output correlation increased with firing rate, even when input correlation was constant.
• Output correlation was largely independent of spike train variability.
• The observed relationship between output correlation and firing rate proved robust to input heterogeneities.

Conclusions:

• Neuronal firing rate, rather than input variability, is a key determinant of correlated spiking.
• This study reveals a significant, previously overlooked relationship between firing rate and correlation in neural activity.
• The findings link two fundamental aspects of the neural code: spiking rate and correlation, offering a more unified understanding.