Recording Visual Evoked Potentials and Auditory Evoked P300 at 9.4T Static Magnetic Field

Jorge Arrubla, Irene Neuner, David Hahn, Frank Boers, N. Jon ShahPLoS 2013 · DOI

Abstract

Simultaneous EEG-fMRI recording offers superior multimodal insights compared to fMRI alone. While recording EEG at ultra-high static magnetic fields (up to 9.4T) is feasible, it presents significant technical challenges. This study investigates the recording of visual evoked potentials (VEP) and auditory P300 signals within a 9.4T environment. Using a 32-channel MR-compatible system on 26 healthy volunteers, researchers successfully retrieved these event-related potentials (ERPs) after correcting for ballistocardiogram (BCG) artifacts via Independent Component Analysis (ICA). Results indicate that while ERP latencies remain stable, amplitudes are significantly higher at 9.4T, likely due to hardware interactions with the magnetic field rather than altered neural processing.

Introduction: The Challenge of Ultra-High Field EEG

Combining the temporal precision of EEG with the spatial resolution of fMRI is a cornerstone of modern neuroimaging. However, as researchers push toward ultra-high field (UHF) strengths like 9.4T to enhance fMRI quality, the EEG signal becomes increasingly obscured by artifacts. The most prominent is the ballistocardiogram (BCG) artifact, caused by cardiac-induced scalp movements and blood flow within the magnetic field. At 9.4T, these artifacts can be orders of magnitude larger than the microvolt-scale neural signals of interest, such as the P100 (visual) and P300 (auditory) potentials.

Methodology

The study involved 26 healthy participants. EEG data were recorded using a 32-channel MR-compatible system in two conditions: a 0T mock scanner (baseline) and a 9.4T static magnetic field.

  • Visual Task: Subjects viewed 200 flashes to elicit the P100 sensory response.
  • Auditory Task: An oddball paradigm (20% target probability) was used to elicit the P300 cognitive response.
  • Signal Processing: Independent Component Analysis (ICA) was employed to isolate and remove BCG artifacts, a critical step for recovering the underlying ERPs at such high field strengths.

Results and Comparative Analysis

The researchers successfully recovered ERPs in the majority of subjects (21/26 for VEP; 19/26 for P300). A key finding was the preservation of ERP timing; latencies for the P100 and P300 showed no significant difference between 0T and 9.4T environments, suggesting that the magnetic field does not slow down early sensory or cognitive processing.

However, signal amplitudes were markedly higher at 9.4T. For instance, the visual P100 mean amplitude rose from 7.49 µV at 0T to 23.69 µV at 9.4T. This contrast is clearly visible when comparing the baseline recordings to those taken within the UHF environment.

EEG signal and ERP waveform at 0T baseline.
EEG signal and ERP waveform at 9.4T, showing significantly increased amplitude.

Discussion and Conclusion

The observed amplitude increase is attributed to the Hall Effect—the interaction between the magnetic field and moving ions within the EEG electrodes and leads—rather than a change in brain physiology. This "amplitude paradox" highlights the need for careful interpretation of EEG data in UHF environments, as the physical environment amplifies the recorded signal without necessarily reflecting increased neural activity.

The study confirms that cognitive markers like the P300 can be reliably recorded at 9.4T using advanced artifact rejection techniques like ICA. Future work will focus on recording during active fMRI sequences, where gradient-switching artifacts present an even greater challenge to signal integrity.

Summary updated

Abstract

Simultaneous recording of electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) has shown a number of advantages that make this multimodal technique superior to fMRI alone. The feasibility of recording EEG at ultra-high static magnetic field up to 9.4T was recently demonstrated and promises to be implemented soon in fMRI studies at ultra high magnetic fields. Recording visual evoked potentials are expected to be amongst the most simple for simultaneous EEG/fMRI at ultra-high magnetic field due to the easy assessment of the visual cortex. Auditory evoked P300 measurements are of interest since it is believed that they represent the earliest stage of cognitive processing. In this study, we investigate the feasibility of recording visual evoked potentials and auditory evoked P300 in a 9.4T static magnetic field. For this purpose, EEG data were recorded from 26 healthy volunteers inside a 9.4T MR scanner using a 32-channel MR compatible EEG system. Visual stimulation and auditory oddball paradigm were presented in order to elicit evoked related potentials (ERP). Recordings made outside the scanner were performed using the same stimuli and EEG system for comparison purposes. We were able to retrieve visual P100 and auditory P300 evoked potentials at 9.4T static magnetic field after correction of the ballistocardiogram artefact using independent component analysis. The latencies of the ERPs recorded at 9.4T were not different from those recorded at 0T. The amplitudes of ERPs were higher at 9.4T when compared to recordings at 0T. Nevertheless, it seems that the increased amplitudes of the ERPs are due to the effect of the ultra-high field on the EEG recording system rather than alteration in the intrinsic processes that generate the electrophysiological responses.
BibTeX
@inproceedings{arrubla2013recording,
  author = {Jorge Arrubla and Irene Neuner and David Hahn and Frank Boers and N. Jon Shah},
  title = {{Recording Visual Evoked Potentials and Auditory Evoked P300 at 9.4T Static Magnetic Field}},
  booktitle = {PLoS},
  year = {2013},
  doi = {10.1371/journal.pone.0062915},
}
Figure