Objective: The aim of the study was to evaluate the sensitivity and reliability of assessing hemispheric language dominance with functional magnetic resonance imaging (fMRI) using a 'free reversed association task.'
Methods: Thirty-nine healthy subjects (13 dextrals, 13 sinistrals and 13 bimanuals) underwent two repeated fMRI sessions. In the active phases sets of words were presented via headphones, and an associated target item was named. During the baseline phases a standard answer was given after listening to unintelligible stimuli. Data were preprocessed with SPM, and then laterality indices (LI) and reliability coefficients (RC) were calculated.
Results: Extensive frontal, temporal and parietal activations were found. Seventy-eight percent of the subjects showed left-hemispheric dominance, 5% showed right-hemispheric dominance, and 17% had bilateral language representations. The incidence of right-hemispheric language dominance was 4.3 times higher in a left-hander with a handedness quotient (HQ) of -90 than in a right-hander with a HQ of +90. The RC was 0.61 for combined ROIs (global network). Strong correlations were found between the two session LIs (r = 0.95 for the global network).
Conclusion: 'Free reversed association' is a sensitive and reliable task for the determination of individual language lateralization. This suggests that the task may be used in a clinical setting.