Measuring functional connectivity during distinct stages of a cognitive task

Area of Science:

• Neuroimaging
• Cognitive Neuroscience
• Brain Connectivity Analysis

Background:

• Functional brain imaging data is complex and multivariate, offering opportunities to study brain interactions during cognition.
• Existing methods for analyzing inter-regional interactions in functional magnetic resonance imaging (fMRI) data have limitations in modeling dynamic changes during cognitive tasks.

Purpose of the Study:

• To introduce a novel method for characterizing inter-regional functional connectivity in event-related fMRI data.
• To enable the modeling of brain region interactions during distinct stages of a cognitive task.

Main Methods:

• The method utilizes separate covariates within the general linear model (GLM) to model activity evoked during each stage of individual trials.
• Stage-specific beta series are generated by sorting parameter estimates, and correlated beta series indicate functional interaction during specific task stages.

Main Results:

• Validation on sequence-tapping tasks showed increased correlations between motor regions during tasks requiring greater bimanual coordination, aligning with prior studies.
• Application to a delayed recognition task generated distinct functional connectivity maps for component stages, highlighting stage-specific neural network dynamics.

Conclusions:

• The developed method provides a powerful tool for analyzing dynamic functional connectivity in event-related fMRI.
• It allows for novel observations of neural networks within isolated stages of cognitive tasks, advancing our understanding of brain function.