Network-based statistics distinguish anomic and Broca's aphasia.

Area of Science:

• Neuroscience
• Neurolinguistics
• Cognitive Science

Background:

• Aphasia, a common post-stroke impairment, affects speech and language due to left hemisphere damage.
• Neural mechanisms underlying different aphasia types remain incompletely understood.
• Resting-state functional magnetic resonance imaging (rs-fMRI) is crucial for studying brain connectivity.

Purpose of the Study:

• To identify differences in resting-state functional connectivity between anomic and Broca's aphasia.
• To elucidate the distinct neural correlates of these two aphasia subtypes.
• To compare network-based statistic (NBS) with lesion symptom mapping (LSM) for identifying neural correlates.

Main Methods:

• Utilized network-based statistic (NBS) and voxel- and connectome-based lesion symptom mapping (V-, CLSM).
• Included lesion volume as a covariate to control for lesion effects in both NBS and LSM.
• Analyzed resting-state functional connectivity differences between anomic and Broca's aphasia groups.

Main Results:

• NBS identified a bilateral dorsal language stream subnetwork with weaker connections in Broca's aphasia compared to anomic aphasia.
• Complex network measures highlighted specific regions (e.g., right inferior frontal sulcus, bilateral superior temporal gyrus) with intensive interaction.
• CLSM identified differentiating connections when restricted to a hypothesized network, but NBS showed advantages for whole-brain analysis.

Conclusions:

• Novel signatures of resting-state brain network differences were identified between anomic and Broca's aphasia groups.
• A specific subnetwork statistically differentiated the resting-state networks of the two groups, offering advantages over standard CLSM.
• Network-level analyses are valuable for investigating neural correlates of post-stroke language deficits.