Functional connectivity across the human subcortical auditory system using an autoregressive matrix-Gaussian copula...

Area of Science:

• Neuroscience
• Auditory System Research
• Brain Imaging Analysis

Background:

• Subcortical auditory system research heavily relies on invasive animal models due to technical imaging limitations.
• Ultrahigh-field functional magnetic resonance imaging (fMRI) enables noninvasive study of human auditory subcortex features like tonotopy.
• Functional connectivity in human subcortical auditory networks remains underexplored, with ongoing methodological development.

Purpose of the Study:

• To develop and validate a novel method for inferring functional connectivity within the human auditory system.
• To specifically investigate direct connectivity patterns in the ascending auditory pathway using partial correlations.
• To address limitations of existing methods by accounting for non-Gaussianity and temporal autocorrelation in fMRI data.

Main Methods:

• Developed an autoregressive matrix-Gaussian copula graphical model (ARMGCGM) to estimate partial correlations from fMRI data.
• Applied the ARMGCGM to analyze functional connectivity in the human auditory pathway, accounting for temporal autocorrelation.
• Validated results through data splitting and cross-validation to ensure stability and reliability across acquisitions.

Main Results:

• Demonstrated strong positive partial correlations between successive structures in the primary auditory pathway (auditory midbrain, thalamus, auditory cortex).
• Showcased high stability of these connectivity findings across different data splits and cross-validation folds.
• Contrasted with full correlation analysis, which identified non-specific interconnectivity, highlighting the specificity of partial correlation findings.

Conclusions:

• Novel connectivity approaches, like ARMGCGM, can reliably recover unique functional connectivity patterns in the auditory pathway.
• Partial correlation analysis, when appropriately modeled, provides more direct and specific insights into auditory network organization.
• The developed methods offer a reliable tool for noninvasive investigation of human subcortical auditory functional connectivity.