Elucidating relations between fMRI, ECoG, and EEG through a common natural stimulus

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Brain Imaging

Background:

• Human brain mapping utilizes fMRI, ECoG, and EEG, capturing distinct physiological signals.
• Previous signal correlations were established in specific tasks or resting states, often in animals.
• Generalizability of these correlations to complex naturalistic stimuli in humans remained uncertain.

Purpose of the Study:

• To assess the reliability of fMRI, ECoG, and EEG responses to naturalistic audio-visual stimuli.
• To determine the similarity of stimulus-related responses across these neuroimaging methods.
• To investigate signal correlations in naturalistic contexts relevant to human cognition.

Main Methods:

• Presented the same movie clip twice to three subject cohorts (N_EEG=45, N_fMRI=11, N_ECoG=5).
• Investigated stimulus-driven correlations across viewings and between imaging methods (fMRI, ECoG, EEG).
• Analyzed responses across various brain regions and frequency bands, controlling for task-irrelevant confounds.

Main Results:

• All three methods demonstrated similar repeat-reliability across viewings when data were averaged across subjects.
• fMRI signal positively correlated with high-frequency ECoG power in cortical structures.
• fMRI showed positive correlations with low-frequency EEG and ECoG power, and infra-slow EEG voltage fluctuations.

Conclusions:

• fMRI, ECoG, and EEG exhibit comparable reliability for naturalistic stimulus processing.
• Cross-modal correlations between fMRI and ECoG/EEG are robust across different frequency bands and contexts.
• Findings extend previous knowledge to natural stimulus processing, supporting multi-modal brain mapping in humans.