Multimodal integration of M/EEG and f/MRI data in SPM12

We describe the steps involved in analysis of multi-modal, multi-subject human neuroimaging data using the SPM12 free and open source software (https://www.fil.ion.ucl.ac.uk/spm/) and a publically-available dataset organised according to the Brain Imaging Data Structure (BIDS) format (https://openneuro.org/datasets/ds000117/). The dataset contains electroencephalographic (EEG), magnetoencephalographic (MEG), and functional and structural magnetic resonance imaging (MRI) data from 16 subjects who undertook multiple runs of a simple task performed on a large number of famous, unfamiliar and scrambled faces. We demonstrate: 1) batching and scripting of preprocessing of multiple runs/subjects of combined MEG and EEG data, 2) creation of trial-averaged evoked responses, 3) source-reconstruction of the power (induced and evoked) across trials within a time-frequency window around the “N/M170” evoked component, using structural MRI for forward modelling and simultaneous inversion (fusion) of MEG and EEG data, 4) group-based optimisation of spatial priors during M/EEG source reconstruction using fMRI data on the same paradigm, and 5) statistical mapping across subjects of cortical source power increases for faces versus scrambled faces.