Correlation Structure in Micro-ECoG Recordings is Described by Spatially Coherent Components

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Electrocorticography (ECoG) is increasingly utilized due to technological advancements and implantation advantages over intracortical methods.
• Optimizing ECoG electrode geometry is crucial given design flexibility and the need for high-resolution neural recordings.
• Conductive polymer (PEDOT:PSS) microelectrodes offer low impedance at small sizes, ideal for evaluating ECoG granularity.

Purpose of the Study:

• To investigate the optimal electrode pitch for electrocorticography (ECoG) recordings.
• To assess the spatial properties of ECoG signals and their frequency dependence.
• To determine the necessary granularity of ECoG grids for capturing detailed neural activity.

Main Methods:

• Utilized two-dimensional (2D) micro-ECoG grids with varying electrode pitches (0.4 mm, 0.2/0.25 mm) for intra-operative human and acute animal recordings.
• Analyzed spatial signal properties by calculating average electrode correlation as a function of pitch.
• Applied independent component analysis (ICA) to identify underlying spatial patterns and sources of correlated activity.

Main Results:

• Found a significant frequency dependence in the spatial scale of signal correlation.
• Demonstrated that spatial correlation patterns are influenced by multiple extended, time-locked neural sources.
• Confirmed the presence of spatially structured brain activity at sub-millimeter scales via ECoG.

Conclusions:

• Dense micro-ECoG grids are justified for capturing fine-grained neural activity.
• Sub-millimeter electrode spacing in ECoG can resolve spatially structured brain signals.
• Findings support the use of high-granularity ECoG for detailed neural mapping and research.