An exact method to quantify the information transmitted by different mechanisms of correlational coding

Area of Science:

• Computational Neuroscience
• Systems Neuroscience
• Information Theory

Background:

• Correlated neuronal firing is a key factor in neural population coding.
• Previous methods for quantifying its impact were limited by approximations and scope.
• Understanding synergy-redundancy effects is crucial for neural information processing.

Purpose of the Study:

• To develop a novel, exact method for quantifying the impact of correlated neuronal firing on information transmission.
• To generalize previous work to longer time windows and larger neuronal populations.
• To dissect synergy-redundancy effects into distinct contributions.

Main Methods:

• Derivation of a new mathematical framework for information quantification.
• Exact consideration of high-order spike train statistics.
• Application to simultaneously recorded neural data from sensory cortices.

Main Results:

• The new method precisely quantifies information transmission, accounting for correlated activity.
• Synergy-redundancy effects are decomposed into redundant, stimulus-dependent correlational, and stimulus-independent correlational contributions.
• The method is validated on real neural data, including high firing rate recordings.

Conclusions:

• The developed method provides a reliable tool for assessing the role of cross-correlated activity in neural coding.
• It offers a comprehensive approach to understanding information processing in neuronal populations.
• The findings are applicable to various brain regions and recording types, including multi-unit activity.