Intrasubject reproducibility of functional MR imaging activation in language tasks.

Area of Science:

• Neuroimaging
• Cognitive Neuroscience
• Neurolinguistics

Background:

• Functional magnetic resonance imaging (fMRI) is a key tool for mapping brain activity during cognitive tasks.
• Assessing the reproducibility of fMRI activation patterns is crucial for reliable clinical and research applications, particularly in language processing.

Purpose of the Study:

• To evaluate the reproducibility of functional MR imaging (fMRI) activation, specifically volume and laterality, in the inferior frontal and temporoparietal regions.
• To compare the reproducibility of different language tasks for their utility in functional neuroimaging studies.

Main Methods:

• Ten healthy participants underwent fMRI scans during six language tasks, with sessions repeated after a minimum of four weeks.
• Laterality indices (LIs) were calculated using both activation volume and F-statistic methods.
• Reproducibility was quantified using concurrence ratios for activation volume and test-retest correlations for LIs.

Main Results:

• All language tasks yielded reproducible LIs in at least one region of interest.
• Verb generation task showed the highest test-retest correlations for LIs (r = 0.99) in both regions.
• LIs derived from F-statistics were more reproducible than those based on activation volume, with lower reproducibility observed in temporoparietal regions.

Conclusions:

• fMRI can reliably assess hemispheric dominance for language processing by providing reproducible LIs in both inferior frontal and temporoparietal areas.
• Activation volume, particularly in temporoparietal regions, is less reproducible and should be interpreted with caution when assessing language lateralization.