| |
Abstract:
Abstract: Research in patients with focal lesions has
indicated that Broca's area is important for syntactic processing
(Caplan et al., 1985; Zurif et al., 1993). Some recent neuroimaging
studies also have shown Broca's area activation during complex
sentence processing (Stromswold et al., 1996; Caplan et al., 1998,
1999); others implicate Wernicke's area and right-hemisphere
homologs of Broca's and Wernicke's areas as well (Just et al.,
1996). In the present fMRI study, participants (n=8, R-handed,
English speakers) matched auditory sentences of two types, either
simple subject-clefts (e.g., The student lifted the biker) or
complex object-clefts (e.g., It was the biker who the student
lifted), to pictures. Subjects responded by button press to matches
and reaction times were recorded. Thirty-two contiguous 4-mm axial
slices were obtained relative to the AC-PC line using whole-brain
echo-planar imaging. Images were motion-corrected and normalized
onto a common stereotactic space using SPM96. Comparing complex
with simple sentence processing revealed activation in Broca's area
(BA 44), dorsolateral prefrontal cortex (BA 9), Wernicke's area,
including posterior superior and middle temporal gyri (BA 22, 21),
and the angular gyrus (BA 39). These findings suggest that complex
syntactic processing engages both Wernicke's and Broca's areas; we
discuss our findings in the context of task, stimulus modality, and
imaging procedure (PET vs. fMRI) differences across studies.
(Supported by McDonnell-Pew Foundation and NIH grants DC01948,
NS30863.)
|
