Cortical oscillations reflect opponent ensemble dynamics through coordinated multifrequency activity.

Area of Science:

• Neuroscience
• Computational Neuroscience

Background:

• Neural oscillations are often used as proxies for neuronal activity.
• Power in individual frequency bands is typically interpreted as indexing neural circuit engagement.
• However, this relationship is inconsistent across regions and contexts, challenging stable frequency-to-circuit mapping.

Purpose of the Study:

• To investigate the relationship between neural oscillations and neuronal activity.
• To challenge the assumption of stable frequency-to-circuit mapping.
• To identify novel organizational principles linking oscillatory dynamics to circuit states.

Main Methods:

• Analysis of multi-frequency amplitude co-fluctuations in rat medial prefrontal cortex.
• Identification of recurring spectral motifs in opponent pairs.
• Comparison of motif opponency balance with frequency band power models during brain-computer interface learning.

Main Results:

• Glutamatergic population activity is linked to dynamically recurring multi-frequency spectral motifs, not isolated frequency bands.
• Spectral motifs occur in opponent pairs with inverted relationships to population activity.
• Shifts in motif opponency balance better explained learning-related activity changes than frequency band power alone.
• Opponent motifs selectively map onto opponent cell ensembles and enable bidirectional LF P-ensemble mapping.

Conclusions:

• Multi-frequency opponent motifs represent a conserved organizational principle linking neural oscillations to population-level circuit states.
• This challenges the view of individual frequency bands as interpretable functional units.
• Spectral motifs offer a more nuanced understanding of how brain oscillations relate to neural activity and learning.