| |
Abstract:
Motion artifact has limited application of fMRI to speech
production. We describe techniques for removing these artifacts
from cortical activation during overt speech, which we combine with
event-related fMRI to compare overt speech to silent speech with
Broca's area and primary motor cortex as regions of interest. After
training to reduce head motion, subjects performed four speech
paradigms, each in a run of 13 32-sec trials.
Whole-brain-plus-vocal-tract images were collected in a 1.5T GE
scanner using 7 mm sagittal slices, 3.75 x 3.75 mm in-plane
resolution, T2* weighted Gradient Echo EPI pulse sequences,
TR/TE=2000/50ms. Subjects named letters aloud, named letters
silently, generated animal names starting with particular letters
aloud, and generated animal names silently. Registration analyses
revealed small, correctable in-plane rotations. Vocal-tract muscles
produced sharply peaked signals which appeared in some
inferior-posterior brain voxels, but not in the regions of
interest, which were then analyzed using cross correlation
techniques with phase and intensity criteria. Inferior Primary
Motor Cortex was activated during overt but not silent speech.
Broca's response was more complicated. Relative to silent speech,
overt letter naming increased Broca's area activation whereas overt
animal-name generation decreased Broca's activation. Though
surprising, part of this pattern has been reported before using
PET, lending some support to our findings and hence our
methods.
|
