Single-trial EEG–fMRI reveals the dynamics of cognitive function

Abstract

Two major non-invasive techniques in cognitive neuroscience, electroencephalography (EEG) and functional magnetic resonance imaging (fMRI), have complementary advantages with regard to their spatial and temporal resolution. Recent hardware and software developments have made it feasible to acquire EEG and fMRI data simultaneously. We emphasize the potential of simultaneous EEG and fMRI recordings to pursue new strategies in cognitive neuroimaging. Specifically, we propose that, by exploiting the combined spatiotemporal resolution of the methods, the integration of EEG and fMRI recordings on a single-trial level enables the rich temporal dynamics of information processing to be characterized within spatially well-defined neural networks.Introduction Electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) recordings provide complementary advantages with regard to the temporal and spatial resolution of brain activity. fMRI enables brain regions engaged during cognitive processes to be localized with high spatial precision; however, the blood-oxygenationlevel-dependent (BOLD) response is too slow to capture fully the rich temporal dynamics that underlie cognitive processes. By contrast, the EEG signal is directly coupled to neuronal electrical activity and has millisecond precision; however, EEG picks up a blurred spatial mixture of the underlying cortical activity and therefore provides only limited spatial resolution. Accordingly, great hope lies in the integration of EEG and fMRI to achieve both high temporal and high spatial resolution of human brain function [1][2][3][4][5]. Here, we propose that simultaneous EEG and fMRI recordings provide a major improvement that will advance considerably our understanding of how cognitive functions are implemented by the brain. Importantly, simultaneous EEG-fMRI recordings enable the investigation of trial-by-trial fluctuations of brain activity, which reveals important insights into the dynamics of cognitive function.The fundamental assumption behind any integration approach is that the signals recorded in both modalities are at least partly produced by the same neural generators.

Unmixing concurrent EEG-fMRI with parallel independent component analysis

“…Although denoising and inclusion of parametric modulations into the stimulus paradigm (Eichele, et al, 2005), and temporal unmixing of the EEG (Debener, et al, 2005b) solve parts of the problem and make way for refined spatiotemporal mapping, there is still need for improvement of the analysis tools for integration of concurrent recordings (cf. Debener, et al, 2006). One such improvement is to unmix both modalities in parallel at the single-trial level, which follows naturally from the recent work (Calhoun, et al, 2006c;Debener, et al, 2005b;Eichele, et al, 2005) and the reasoning laid out above.…”

“…These neuronal mass responses can be observed across scales and modalities from single unit recordings, intracranial and scalp electrophysiology, as well as metabolic and hemodynamic signals, but no single technique provides a sufficient view of the full temporal, spatial and functional extent of these responses. Visibility can be improved with techniques that integrate data across different neuroimaging modalities (Debener, et al, 2006;Hopfinger, et al, 2005;Horwitz, et al, 2002;Makeig, 2002). In the case of concurrent EEG-fMRI recordings, one can complement the temporal resolution provided by scalp potentials with the spatial precision of fMRI.…”

Frontal theta EEG activity correlates negatively with the default mode network in resting state

“…Previous studies (Makeig et al, 2002;Onton et al, 2005) have demonstrated that ICA can successfully be applied to EEG data to separate frontal theta activity from other (oscillatory) activity. Debener et al (2005Debener et al ( , 2006 demonstrated that ICA can be applied to EEG data measured in the MRI scanner to obtain single trial measures of evoked potentials that can be successfully correlated with the BOLD signal. Similarly, by correlating power fluctuations of the frontal theta components obtained by ICA with the BOLD signal in resting state, we hope to gain some insight into the neuronal structures that are associated with the frontal theta rhythm in humans.…”

Detection of experimental ERP effects in combined EEG–fMRI: Evaluating the benefits of interleaved acquisition and Independent Component Analysis

“…In statistical terms, we set the goal of assessing whether the experimental contrast ('go' vs. 'nogo')-to-noise ratio was greater in the ERPs acquired in the absence of concurrent MR gradient switching than in the ERPs acquired with concurrent gradient switching, both in averaged ERPs and over individual EEG segments. The latter analysis was motivated by the increasingly popular approach of correlating ERP component amplitude (or latency) with the fMRI signal over single trials (see Debener et al, 2006, for a review).…”

“…Indeed, correlating the two measurements over single trials (e.g. Debener et al, 2005;Eichele et al, 2005;Bénar et al, 2007;Mantini et al, 2009;Warbrick et al, 2009) has been one of the primary motivations for combining EEG and fMRI acquisition (see Debener et al, 2006, andLaufs et al, 2008, for reviews).…”