This study investigated the influence of large-scale neural dynamics, measured using dynamic functional connectivity (dFC), on treatment-induced functional network changes in patients with post-stroke aphasia (PSA).

Recovery from PSA varies greatly between individuals. While factors such as initial severity and lesion characteristics are among the most reliable predictors, variation in recovery trajectories remains incompletely explained. Previous work investigating dFC (i.e., time-varying interregional correlations in the fMRI BOLD signal) in PSA found that greater temporal variability (TV) of dFC predicted greater treatment-induced improvement in picture naming. We proposed the following mechanism for this relationship: (1) Transient inter-regional synchronization facilitates synaptic plasticity between regions, and (2) greater TV represents a greater diversity of transient connectivity configurations sampled over time, creating more diverse opportunities for plasticity.

Pre- and post-treatment resting-state static functional connectivity (sFC, i.e., time-invariant interregional fMRI BOLD signal correlations) was assessed in 30 participants with PSA who received a semantic treatment for aphasia. Network property changes were quantified using the graph metrics, node strength, which measures each region's connection strength to the rest of the cortex, and small-worldness, which measures the balance between dense local clustering and global network efficiency. These changes were then related to (1) treatment-induced behavioral improvement and (2) baseline TV of dFC using linear regression models.

Higher TV was predictive of treatment-induced decreases in node strength, which were in turn associated with greater behavioral treatment gains. These decreases in node-level strength were also associated with global increases in small-worldness. Additionally, increases in small-worldness were associated with better behavioral treatment response.

These findings support the hypothesis that higher TV facilitates node-level network changes, resulting in global network alterations that promote recovery in PSA.