Sparse Multivariate Analysis Reveals Dissociable White Matter Networks for Cognitive and Motor Processing Speed

Area of Science:

• Cognitive Neuroscience
• Neuroimaging
• Brain Mapping

Background:

• Reaction time (RT) measures information processing speed, influenced by brain structure and function.
• Previous research linked white matter and EEG alpha oscillations to cognition separately.
• The combined influence on distinct RT aspects remains unclear.

Purpose of the Study:

• Investigate multimodal data's role in dissociating neural systems for cognitive and motor processing speed.
• Examine associations between white matter tracts, EEG alpha frequency, and RT metrics.
• Utilize sparse canonical correlations to find multivariate links across modalities.

Main Methods:

• Diffusion tensor imaging (DTI) and resting-state EEG data from 24 healthy adults.
• GO/NO-GO paradigm for behavioral RT (mean, standard deviation, skewness).
• Sparse multiple canonical correlation analysis (SMCCA) for cross-modal associations.

Main Results:

• Two dimensions revealed: fronto-temporal tracts linked to complex RT (cognitive processing).
• Motor/interhemispheric tracts associated with RT skewness (motor consistency).
• Individual alpha peak frequency (IAF) showed minimal contribution; sex was strongly associated.

Conclusions:

• Separate white matter networks underpin distinct cognitive and motor RT aspects.
• Resting-state alpha frequency did not consistently link to behavioral variability in this sample.
• Multimodal, multivariate approaches are crucial for complex brain-behavior relationships.