Temporal dynamics of energy-efficient coding in mouse primary visual cortex

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Sparse coding is a key principle for efficient sensory information representation in the brain.
• The temporal dynamics of sparse coding and its impact on neural representations are not fully understood.

Purpose of the Study:

• To investigate the temporal dynamics of sparse coding in cortical populations during sensory processing.
• To understand how coding efficiency changes over time following stimulus onset.

Main Methods:

• Analysis of cortical population activity in response to sensory stimuli.
• Quantification of sparseness, mutual information, and metabolic cost.
• Examination of the relationship between these metrics over time.

Main Results:

• Stimulus onset initially broadens cortical activation, decreasing sparseness and increasing mutual information.
• Competitive neural interactions maintain mutual information as activity decreases and sparseness recovers.
• Coding efficiency, defined as mutual information relative to metabolic cost, progressively increases.

Conclusions:

• Cortical sensory representations are dynamically optimized over time.
• The brain balances information transmission with metabolic constraints for efficient neural coding.