Correlation structure of grid cells is preserved during sleep

Area of Science:

• Neuroscience
• Computational Neuroscience
• Spatial Navigation

Background:

• Grid cells in the medial entorhinal cortex (MEC) are crucial for spatial mapping.
• Continuous attractor network models propose mechanisms for grid cell activity but lack experimental validation regarding state independence.
• The correlation structure of grid cell networks across different behavioral states remains largely unexplored.

Purpose of the Study:

• To investigate whether the correlation structure of medial entorhinal cortex (MEC) grid cell activity is independent of behavioral states.
• To test the prediction of continuous attractor network models regarding state-invariant correlations in grid cell ensembles.
• To explore the functional independence of different grid cell modules.

Main Methods:

• Recorded ensemble activity of MEC cells in rodents during both navigation (awake behavior) and sleep.
• Analyzed spike rate correlations and spatial phase offsets of grid cells.
• Examined correlations between conjunctive grid-head direction cells and pure head direction cells.
• Assessed correlations across different grid modules and their scale relationships during various brain states.

Main Results:

• Spatial phase offsets of grid cells predicted arousal-state-independent spike rate correlations.
• State-invariant correlations between conjunctive and head direction cells were predicted by head direction tuning offsets during awake behavior.
• Weak correlations were observed across grid modules, and module scale relationships were disrupted during slow-wave sleep.
• Grid modules appear to function as independent attractor networks.

Conclusions:

• The network states in the medial entorhinal cortex (MEC) are expressed universally across different brain and behavioral states.
• Findings support the hypothesis that grid cell network correlations are largely independent of behavioral context.
• Grid modules likely operate as distinct, independent attractor networks.