hide
Free keywords:
-
Abstract:
Most neuroimaging studies are performed using one imaging method only, either functional
magnetic resonance imaging (fMRI), electroencephalography (EEG), or magnetoencephalography
(MEG). Information on both location and timing has been sought by recording fMRI and EEG, simultaneously,
or MEG and fMRI in separate sessions. Such approaches assume similar active areas whether
detected via hemodynamic or electrophysiological signatures. Direct comparisons, after independent
analysis of data from each imaging modality, have been conducted primarily on low-level sensory
processing. Here, we report MEG (timing and location) and fMRI (location) results in 11 subjects when
they named pictures that depicted an action or an object. The experimental design was exactly the
same for the two imaging modalities. The MEG data were analyzed with two standard approaches: a
set of equivalent current dipoles and a distributed minimum norm estimate. The fMRI blood-oxygenlevel
dependent (BOLD) data were subjected to the usual random-effect contrast analysis. At the group
level, MEG and fMRI data showed fairly good convergence, with both overall activation patterns and
task effects localizing to comparable cortical regions. There were some systematic discrepancies, however,
and the correspondence was less compelling in the individual subjects. The present analysis
should be helpful in reconciling results of fMRI and MEG studies on high-level cognitive functions
