Spatial and Temporal Resolution of fMRI and HD EEG

The temporal resolution of EEG is well known to researchers and clinicians; EEG directly measures neuronal activity. On the other hand, it is commonly believed that EEG provides poor spatial detail, due to the fact the EEG signal is recorded at a distance from the source generator, the signals are “distorted” by the inhomogeneous conductivity properties of different head tissues, and the ill-posed nature of the source-estimate inverse problem.

However, given advances in dense-array EEG recordings, image processing, computational power, and inverse techniques, it is time to re-evaluate this common assumption of spatial resolution. The EGI Science Team has been at the forefront of developments in dense-array EEG technology, MR image segmentation, and big-compute solutions that enable us to address the current, practical spatial resolution of dense-array EEG for neuroimaging.

Figure 1. Location of peak motor-related activity for fMRI (black star) and event-related spectral changes (high-gamma: red triangle; low-gamma: white diamond; beta: brown crescent; mu: purple circle).

Recent motor and somatosensory activation studies, conducted by EGI scientists and performed with both HD EEG and fMRI, show that when accurate head models are employed, there is good correspondence between blood oxygen level dependent (BOLD) activity and EEG spectral changes (before actual finger movement, see Figure 1). As can be seen in Figure 1, the majority of spectral related changes are localized to the primary motor cortex of the hand region (green area), similar to the peak BOLD location.

Figure 2. Left: peak source activity for EEG P50 component from lateral and front views. Right: Peak BOLD response from lateral and front views.

For the somatosensory cortex, activity approximately 50 ms after stimulation of the thumb (same digit used for the motor study in Figure 1) is clearly localized to the anterior and posterior banks of the central sulcus (Figure 2) at the level of primary motor and somatosensory cortex (yellow area in Figure 1). The HD EEG findings are consistent with anatomical evidence: inputs from the somatosensory nucleus of the thalamus project to the sulcal banks of the primary somatosensory cortex (Kass 1993) and somatosensory responses span both the anterior and posterior banks(Tanji and Wise, 1981). In contrast, the peak BOLD response is in the secondary somatosensory cortex.

References

Kass, J. H. (1993). The functional organization of somatosensory cortex in primates. Annals of Anatomy, 175, 509-518.

Tanji, J., and Wise, S. P. (1981). Submodality distribution in sensorimotor cortex of the unanesthetized monkey. J Neurophysiol, 45, 467-481.