This review outlines the rationale for the use of magnetoencephalography (M
EG) or magnetic source imaging (MSI), a noninvasive functional imaging tech
nique, and the features that any imaging method should display to make a su
bstantial contribution to cognitive neuroscience. After a brief discussion
of the basic experimental approach used in the authors' studies, the use of
early sensory components of brain magnetic responses is reviewed to addres
s issues of the functional organization of the primary sensory cortices, fo
llowed by a comment on the clinical use of these components. Second, normat
ive studies focusing on the late components of magnetic responses for estab
lishing the validity and reliability of MSI maps of the language-specific c
ortex in normal subjects are reviewed. Third, the authors' investigations o
f fine spatiotemporal features of brain activation maps, specific to recept
ive language and to reading, are reviewed. Fourth, experience with presurgi
cal mapping of the language-specific cortex in neurosurgery candidates and
in patients undergoing the "Wada" procedure is summarized followed by a com
ment on the perfect agreement of the MSI maps with those derived by more di
rect invasive brain mapping procedures. Fifth, MSI-derived evidence of ofte
n dramatic, functional reorganization of brain areas subserving both simple
sensory and linguistic functions is summarized along with comments on the
use of MSI as a means for investigating brain plasticity. Finally, in the s
ixth section of this review, the authors relate their experience with the u
se of MSI in deriving brain activation profiles during silent reading of re
al words and pseudowords that are specific to dyslexic children. The review
concludes with a discussion on the further use of MSI in assessing, among
other issues, the effectiveness of intervention strategies designed to impr
ove reading fluency in dyslexic children.