AUTHOR=Schmüser Lena , Sebastian Alexandra , Mobascher Arian , Lieb Klaus , Tüscher Oliver , Feige Bernd TITLE=Data-driven analysis of simultaneous EEG/fMRI using an ICA approach JOURNAL=Frontiers in Neuroscience VOLUME=8 YEAR=2014 URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2014.00175 DOI=10.3389/fnins.2014.00175 ISSN=1662-453X ABSTRACT=

Due to its millisecond-scale temporal resolution, EEG allows to assess neural correlates with precisely defined temporal relationship relative to a given event. This knowledge is generally lacking in data from functional magnetic resonance imaging (fMRI) which has a temporal resolution on the scale of seconds so that possibilities to combine the two modalities are sought. Previous applications combining event-related potentials (ERPs) with simultaneous fMRI BOLD generally aimed at measuring known ERP components in single trials and correlate the resulting time series with the fMRI BOLD signal. While it is a valuable first step, this procedure cannot guarantee that variability of the chosen ERP component is specific for the targeted neurophysiological process on the group and single subject level. Here we introduce a newly developed data-driven analysis procedure that automatically selects task-specific electrophysiological independent components (ICs). We used single-trial simultaneous EEG/fMRI analysis of a visual Go/Nogo task to assess inhibition-related EEG components, their trial-to-trial amplitude variability, and the relationship between this variability and the fMRI. Single-trial EEG/fMRI analysis within a subgroup of 22 participants revealed positive correlations of fMRI BOLD signal with EEG-derived regressors in fronto-striatal regions which were more pronounced in an early compared to a late phase of task execution. In sum, selecting Nogo-related ICs in an automated, single subject procedure reveals fMRI-BOLD responses correlated to different phases of task execution. Furthermore, to illustrate utility and generalizability of the method beyond detecting the presence or absence of reliable inhibitory components in the EEG, we show that the IC selection can be extended to other events in the same dataset, e.g., the visual responses.