Temporal coupling of field potentials and action potentials in the neocortex.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Signal Processing

Background:

• The local field potential (LFP) reflects aggregate neuronal activity and is linked to single-neuron action potentials.
• Increased action potential firing rates are associated with high-frequency band oscillations (50-200 Hz).
• The contribution of action potentials to the LFP signal complicates the direct interpretation of the spike-LFP relationship.

Purpose of the Study:

• To investigate the relationship between neuronal spike rates and LFP across various frequency bands in rat neocortex.
• To identify which LFP frequency bands most consistently correlate with neuronal firing.
• To develop an improved method for predicting action potential activity from LFP signals.

Main Methods:

• Analysis of rat neocortical recordings.
• Examination of correlations between spike rates and LFP power in different frequency bands.
• Investigation of the contribution of spiking and electromyographic activity to LFP power.
• Development of an oscillation-based predictor using the full LFP frequency spectrum.

Main Results:

• Oscillations in the 50-180 Hz range showed the strongest correlation with high neuronal firing rates.
• Other LFP bands also contained information about spiking, including inverse correlations.
• Neuronal spiking and electromyographic activity were found to contribute to LFP power.
• The relationship between spike rates and LFP power varied across different brain states and individual neurons.
• An improved predictor of action potential activity was developed by incorporating information from the entire LFP frequency spectrum.

Conclusions:

• The 50-180 Hz LFP band is a key indicator of neuronal firing, but not the sole determinant.
• Spiking and other physiological activities influence the LFP signal, adding complexity.
• Utilizing the full LFP frequency spectrum enhances the prediction of neuronal spiking activity.
• Understanding these complexities is crucial for accurately inferring neural activity from LFP recordings.