Does the default-mode functional connectivity of the brain correlate with working-memory performances?

The "default-mode" network is an ensemble of cortical regions that are typically deactivated during demanding cognitive tasks in functional magnetic resonance imaging (fMRI) studies. Using functional connectivity analysis, this network can be studied as a "stand-alone" brain system whose functional role is supposed to consist in the dynamic control of intrinsic processing activities like attention focusing and task-unrelated thought generation and suppression. Independent component analysis (ICA) is the method of choice for generating a statistical image of the "default-mode" network (DMN) using a task- and seed-independent distributed model of fMRI functional connectivity without prior specification of node region extent and timing of neural activation. We used a standard graded working-memory task (n-back) to induce fMRI changes in the default-mode regions and ICA to evaluate to DMN functional connectivity in nineteen healthy volunteers. Based on the known spatial variability of the ICA-DMN maps with the task difficulty levels, we hypothesized the ICA-DMN may also correlate with the subject performances. We confirmed that the relative extent of the anterior and posterior midline spots within the DMN were oppositely (resp. positively in the anterior and negatively in the posterior cingulate cortex) correlated with the level of task difficulty and found out that the spatial distribution of DMN also correlates with the individual task performances. We conclude that the working-memory function is related to a spatial re-configuration of the DMN functional connectivity, and that the relative involvement of the cingulate regions within the DMN might function as a novel predictor of the working-memory efficiency.