Brain activation modulated by sentence comprehension

Science

Volumn 274, Issue 5284, 1996, Pages 114-116

  Just, Marcel Adam ^a   Carpenter, Patricia A ^a   Keller, Timothy A ^a   Eddy, William F ^a   Thulborn, Keith R ^b  

^a CARNEGIE MELLON UNIVERSITY   (United States)

^b UNIVERSITY OF PITTSBURGH MEDICAL CENTER   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ARTICLE; COGNITION; COMPREHENSION; ELECTROENCEPHALOGRAM; FRONTAL CORTEX; LEFT HEMISPHERE; NERVE CONDUCTION; NEUROANATOMY; NUCLEAR MAGNETIC RESONANCE IMAGING; PRIORITY JOURNAL; TEMPORAL CORTEX;

ADOLESCENT; ADULT; BRAIN MAPPING; COGNITION; DOMINANCE, CEREBRAL; FEMALE; FRONTAL LOBE; HUMANS; LANGUAGE TESTS; MAGNETIC RESONANCE IMAGING; MALE; TEMPORAL LOBE;

EID: 0029745428     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.274.5284.114     Document Type: Article

References (29)

1. M. A. Just and P. A. Carpenter, Psychol. Rev. 99, 122 (1992).
2. Other graded tasks include studies of short-term list memory [P. M. Grasby et al., Brain 117, 1271 (1994); T. S. Braver et al., Soc. Neurosci. Abstr. 21, 274 (1995)].
3. Other graded tasks include studies of short-term list memory [P. M. Grasby et al., Brain 117, 1271 (1994); T. S. Braver et al., Soc. Neurosci. Abstr. 21, 274 (1995)].
4. S. G. Baker et al., Neuropsychologia 34, 515 (1996).
5. R. J. Binaer et al., Ann. Neurol. 35, 662 (1994); B. M. Mazoyer et al., J. Cogn. Neurosci. 5, 467 (1993).
6. R. J. Binaer et al., Ann. Neurol. 35, 662 (1994); B. M. Mazoyer et al., J. Cogn. Neurosci. 5, 467 (1993).
7. K. Stromswold, D. Caplan, N. Alpert, S. Rauch, Brain Lang. 52, 452 (1996).
8. G. Bottini et al., Brain 117, 1241 (1994).
9. M. Ford, J. Verbal Learn. Verbal Behav. 22, 203 (1983); J. King and M. A. Just, J. Mem. Lang. 30, 580 (1991); W. A. Stolz, J. Verbal Learn. Verbal Behav. 6, 867 (1967); M. A. Just and P. A. Carpenter, Can. J. Exp. Psychol. 47, 310 (1993).
10. M. Ford, J. Verbal Learn. Verbal Behav. 22, 203 (1983); J. King and M. A. Just, J. Mem. Lang. 30, 580 (1991); W. A. Stolz, J. Verbal Learn. Verbal Behav. 6, 867 (1967); M. A. Just and P. A. Carpenter, Can. J. Exp. Psychol. 47, 310 (1993).
11. M. Ford, J. Verbal Learn. Verbal Behav. 22, 203 (1983); J. King and M. A. Just, J. Mem. Lang. 30, 580 (1991); W. A. Stolz, J. Verbal Learn. Verbal Behav. 6, 867 (1967); M. A. Just and P. A. Carpenter, Can. J. Exp. Psychol. 47, 310 (1993).
12. M. Ford, J. Verbal Learn. Verbal Behav. 22, 203 (1983); J. King and M. A. Just, J. Mem. Lang. 30, 580 (1991); W. A. Stolz, J. Verbal Learn. Verbal Behav. 6, 867 (1967); M. A. Just and P. A. Carpenter, Can. J. Exp. Psychol. 47, 310 (1993).
13. H. H. Haarmann, M. A. Just, P. A. Carpenter, Brain Lang., in press.
14. Informed consent with a protocol approved by the University of Pittsburgh Institutional Review Board was obtained from 10 male and 5 female right-handed native English speakers, who were college students or graduates, aged 18 to 30.
15. The gradient echo, resonant echo-planar fMRI used BOLD (blood oxygen level-dependent) contrast [S. Ogawa et al., Biophysical J. 64, 803 (1993)] in a 1.5-T GE Medical Systems scanner at the Magnetic Resonance Research Center of the University of Pittsburgh Medical Center, Images were acquired at a repetition time (TR) of 1500 ms in each of seven adjacent axial planes covering the superior and medial portions of the temporal lobe, with time to echo (TE) of = 50 ms, flip angle = 90°, and voxel sizes of 3.125 mm by 3.125 mm by 5 mm. The surface coils were 12.7-cm bitemporal general purpose coils for 12 of the 15 participants, a left temporal coil for two participants, and a whole head coil for one. In the ancillary experiment, seven coronal slices covered the frontal lobes. In both experiments, the field of view was 400 mm by 200 mm, a 128 by 64 acquisition matrix, 5-mm slice thickness, and a 1 -mm gap. The structural images for anatomic localization were high-resolution T1 - weighted spin echo images, TR = 400 ms, TE = 11 ms, 256 by 256 acquisition matrix, 5-mm slice thickness, and a 1 -mm gap.
16. 5 for a 61 -image comparison and is more conservative than the Bonferroni correction for P < 0.01. The activation effects are believed to reflect changes in the microvasculature induced by changes in cortical neuronal activity rather than in larger veins because (i) regions of activation map to gray matter rather than to spaces occupied by cerebrospinal fluid in which such vessels are found; (ii) the same effects were observed with one of the participants who was tested a second time at 3 T with an asymmetric spin echo pulse sequence that is less sensitive to large vessels [R.M. Weisskoff, C. S. Zuo, J. L. Boxerman, B. R. Rosen, Magn. Reson. Med. 31, 601 (1994)]; (iii) the nonlinear distribution of activation confined to a discrete volume within the imaging space does not represent the known venous drainage pattern of the temporal lobe; and (iv) we excluded from analysis 1.9% of voxels showing greater than 6% change in signal intensity that might have possibly arisen from blood vessels Finally, the problem of blood vessel artifacts in fMRI primarily affects the absolute localization of function, rather than the measurement of the relative volume of activation across conditions. Variations in epoch durations in the three sentence conditions had little effect on the results, as indicted by a separate analysis that truncated each participant's epochs to the duration of the shortest epoch type.
17. 5 for a 61 -image comparison and is more conservative than the Bonferroni correction for P < 0.01. The activation effects are believed to reflect changes in the microvasculature induced by changes in cortical neuronal activity rather than in larger veins because (i) regions of activation map to gray matter rather than to spaces occupied by cerebrospinal fluid in which such vessels are found; (ii) the same effects were observed with one of the participants who was tested a second time at 3 T with an asymmetric spin echo pulse sequence that is less sensitive to large vessels [R.M. Weisskoff, C. S. Zuo, J. L. Boxerman, B. R. Rosen, Magn. Reson. Med. 31, 601 (1994)]; (iii) the nonlinear distribution of activation confined to a discrete volume within the imaging space does not represent the known venous drainage pattern of the temporal lobe; and (iv) we excluded from analysis 1.9% of voxels showing greater than 6% change in signal intensity that might have possibly arisen from blood vessels Finally, the problem of blood vessel artifacts in fMRI primarily affects the absolute localization of function, rather than the measurement of the relative volume of activation across conditions. Variations in epoch durations in the three sentence conditions had little effect on the results, as indicted by a separate analysis that truncated each participant's epochs to the duration of the shortest epoch type.
18. 5 for a 61 -image comparison and is more conservative than the Bonferroni correction for P < 0.01. The activation effects are believed to reflect changes in the microvasculature induced by changes in cortical neuronal activity rather than in larger veins because (i) regions of activation map to gray matter rather than to spaces occupied by cerebrospinal fluid in which such vessels are found; (ii) the same effects were observed with one of the participants who was tested a second time at 3 T with an asymmetric spin echo pulse sequence that is less sensitive to large vessels [R.M. Weisskoff, C. S. Zuo, J. L. Boxerman, B. R. Rosen, Magn. Reson. Med. 31, 601 (1994)]; (iii) the nonlinear distribution of activation confined to a discrete volume within the imaging space does not represent the known venous drainage pattern of the temporal lobe; and (iv) we excluded from analysis 1.9% of voxels showing greater than 6% change in signal intensity that might have possibly arisen from blood vessels Finally, the problem of blood vessel artifacts in fMRI primarily affects the absolute localization of function, rather than the measurement of the relative volume of activation across conditions. Variations in epoch durations in the three sentence conditions had little effect on the results, as indicted by a separate analysis that truncated each participant's epochs to the duration of the shortest epoch type.
19. The magnitude of the activation response for the activated voxels in this set was approximately a 2% change in signal intensity above the rest condition, which is the usual magnitude in many fMRI studies of the association cortex at 1.5 T. There was some deactivation - an average of four voxels per participant in the left temporal ROI (compared with the rest condition) - although its interpretation is unclear [J. Sergent, Trends Neurosci. 17, 221 (1994)].
20. Four ROIs were anatomically defined a priori for each participant with reference to coregistered structural images in the axial and in the sagittal planes. The temporal ROIs were three adjacent axial slices (subtending 17 mm) that best covered the laterosuperior and middle temporal gyn. The frontal ROIs were three adjacent coronal slices that best covered the inferior frontal gyrus. Both ROIs were bounded in-plane by the limiting sulci of the gyn. Anatomical localization of Wernicke's area activation in one participant, scanned a second time at 3 T, indicated activation in both the superior temporal gyrus and sulcus, with Talairach coordinates spanning the ranges: x, -50 to -65; y, -21 to -44; and z, 3 to 19.
21. E. Paulesu, C, D. Frith, R. S. J. Frackowiak, Nature 362, 342 (1993).
22. H. Karbe et al., Neurology 39, 1083 (1989); D. Kempler, S. Curtiss, E. J. Metter, C. A. Jackson, J. Neurolinguist. 6, 301 (1991); E. J. Metter et al., Arch. Neurol. 47, 1235 (1990).
23. H. Karbe et al., Neurology 39, 1083 (1989); D. Kempler, S. Curtiss, E. J. Metter, C. A. Jackson, J. Neurolinguist. 6, 301 (1991); E. J. Metter et al., Arch. Neurol. 47, 1235 (1990).
24. H. Karbe et al., Neurology 39, 1083 (1989); D. Kempler, S. Curtiss, E. J. Metter, C. A. Jackson, J. Neurolinguist. 6, 301 (1991); E. J. Metter et al., Arch. Neurol. 47, 1235 (1990).
25. L. Rueckert et al., J. Neuroimaging 4, 67 (1994).
26. C. A. Tompkins and C. A. Mateer, Brain Lang. 24, 185 (1985).
27. J. D. Cohen et al., Hum. Brain Imaging 1, 293 (1994); J. Jonides et al., Nature 363, 623 (1993).
28. J. D. Cohen et al., Hum. Brain Imaging 1, 293 (1994); J. Jonides et al., Nature 363, 623 (1993).
29. This research was funded in part by Office of Naval Research grant N00014-92-J-1209. National Institute of Mental Health Grant MH-29617, Research Scientist Development Awards MH-00661 and MH-00662, and National Institute of Mental Health training grant MH-19102, and the statistics were funded by NSF grants IBN-9418982 and DMS-9505007 to W.E. We thank T. Egan, P. Koseff, A. Mockus, B. Organ, and J. Sweeney.