Different Approximation Methods for Calculation of Integrated Information Coefficient in the Brain during...

Area of Science:

• Neuroscience
• Computational Neuroscience
• Information Theory

Background:

• Integrated Information Theory (IIT) uses Φ to measure brain adaptation.
• Calculating Φ for large neural datasets is computationally intensive.
• Previous Φ calculations were limited to averaged neural activity or small systems.

Purpose of the Study:

• To develop fast and precise methods for calculating Φ from neural spike data.
• To assess the capability of Φ in describing brain network adaptation.
• To investigate neural plasticity during learning using Φ.

Main Methods:

• Applied approximation methods for Φ calculation on time-series neural spike data.
• Recorded neural activity from rat hippocampus during instrumental learning.
• Correlated Φ values with behavioral performance (successful acts).

Main Results:

• Approximation methods accurately reflected temporal trends of Φ during learning.
• Φ positively correlated with the number of successful actions.
• A specific subgroup of neurons showed modulated Φ during learning.

Conclusions:

• Fast Φ calculation methods enable studying brain adaptation in complex neural networks.
• Φ can serve as a measure of neural plasticity during task acquisition.
• Findings support the application of IIT in understanding learning and adaptation.