Abstract

MagnetoEncephaloGraphy (MEG) relies on the detection of cortical current flow by measurement of the associated magnetic field outside the head. The amplitude of this magnetic field depends strongly on the depth of the electrical brain activity. Additionally, radially orientated sources are magnetically silent in a concentrically homogeneous volume conductor, giving rise to the anecdotal assumptions that MEG is insensitive to both deep and gyral sources. Utilising cortical surfaces extracted from Magnetic Resonance Images (MRIs) of two adult brains we constructed all possible single source elements and examined the proportion of active neocortex that is actually detectable with a whole-head MEG system. We identified those electrically active regions to which MEG is maximally sensitive by analytically computing the probability of detecting a source within a specified confidence volume. Our findings show that source depth, and not orientation, is the main factor that compromises the sensitivity of MEG to activity in the adult human cortex. There are thin strips (∼2 mm wide) of poor resolvability at the crests of gyri; however, these strips account for only a relatively small proportion of the cortical area and are abutted by elements with nominal tangential component yet high resolvability due to their proximity to the sensor array. Finally, we varied the extent of the patches of cortical activity, showing that small patches have a small net-current moment and are therefore less visible whereas large patches have a strong net-current moment, are generally more visible to the MEG system, yet are less appropriately modelled as single dipoles.