A Magnetoencephalographic (meg) Study of Language Lateralization

Abstract:
Abstract: In this experiment we sought to develop and validate a noninvasive, MEG method of determining hemispheric language dominance. The study was divided into two phases. In phase I, auditory evoked magnetic responses were recorded successfully from 71 normal volunteers (51 right-handed, 20 left-handed) using a whole-head biomagnetometer. Responses were evoked using a dichotic listening paradigm, with the subjects evaluating the semantic relatedness of different words presented to the right and left ears. In phase II, noninvasive MEG inferences of language dominance were validated with respect to data from Wada testing in 35 neurosurgical patients (26 right-handed, 9 left-handed). In phase I, the MEG data indicated that 88% of the right-handed subjects had left hemisphere language dominance, 6% bilateral language representation, and 6% right-hemisphere dominance. In left-handed subjects, 60% showed left-hemisphere dominance, 10% bilateral language representation, and 30% right hemisphere dominance. In phase II, the MEG-based assessment of cerebral dominance for language matched that derived from the Wada test in 33 of the 35 patients. MEG testing correctly predicted all cases of mixed or right hemisphere language dominance. MEG is an effective strategy for noninvasive specification of cerebral dominance for language in control subjects and neurosurgical patients.