Abstract
Non-invasive recordings of human brain activity through electroencephalography (EEG) or magnetoencelphalography (MEG) are of value for both basic science and clinical applications in sensory, cognitive, and affective neuroscience. Here we introduce a new approach to estimating the intra-cranial sources of EEG/MEG activity measured from extra-cranial sensors. The approach is based on the group lasso, a sparse-prior inverse that has been adapted to take advantage of functionally-defined regions of interest for the definition of physiologically meaningful groups within a functionally-based common space. Detailed simulations using realistic source-geometries and data from a human Visual Evoked Potential experiment demonstrate that the group-lasso method has improved performance over traditional ℓ2 minimum-norm methods. In addition, we show that pooling source estimates across subjects over functionally defined regions of interest results in improvements in the accuracy of source estimates for both the group-lasso and minimum-norm approaches.
Highlights
Non-invasive recordings of human brain activity through electroencephalography (EEG) or magnetoencelphalography (MEG) provide high-temporal resolution measures of neural activity
We show that group lasso inversion, operating on functional Regions of Interest (ROIs), improves source recovery above and beyond what can be accomplished with the classical minimum norm for single subjects
We evaluate the methods on multiple subjects, and demonstrate that the effectiveness of the group lasso increases with the number of subjects
Summary
Non-invasive recordings of human brain activity through electroencephalography (EEG) or magnetoencelphalography (MEG) provide high-temporal resolution measures of neural activity. When combined with inverse modeling techniques, they provide information about the underlying distribution of neural activity. Starting in the 1990’s, distributed inverse solutions based on the minimum l2 norm approach ( known as ridge regression) began to appear [3,4,5,6]. These methods model the underlying source distribution as a large set of elementary currents, either distributed throughout the intra-cranial volume, or constrained to gray matter. The l2 penalty is based on source power: many weakly activated
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
