Assessing functional connectivity of neural ensembles using directed information.

Area of Science:

• Neuroscience
• Information Theory
• Computational Biology

Background:

• Neural networks are complex, with functional connectivity traditionally assessed using correlation, which lacks directionality.
• Directed information theory offers a potential method for inferring directional influence between neurons.
• The added value of directed information over correlation for neural data remains unclear.

Purpose of the Study:

• To present a modified procedure for estimating directed information.
• To compare the efficacy of directed information with correlation analyses for neural data.
• To explore the utility of directed information in understanding neural ensemble functional connectivity.

Main Methods:

• Development of a modified directed information estimation procedure.
• Application of the method to simulated neural spike train data.
• Analysis of experimental neurophysiological data from rodent and primate studies.

Main Results:

• Directed information outperforms correlation in identifying neural connections and their directionality.
• The method accurately estimates conduction delays between brain structures in experimental data.
• Directed information uncovers connectivity patterns not detectable by correlation analyses.

Conclusions:

• Directed information provides accurate and novel insights into functional neural connectivity.
• This approach enhances the analysis of neurophysiological data from awake, behaving animals.
• Directed information offers a significant advancement over traditional correlation methods for neural network analysis.