Fig. 7.

Comparison of spatial normalisation between FEAT and dHCP pipelines using t-statistics from the GLM output for all 15 subjects. The significant activity in the thresholded maps (TFCE, default parameters, 5% FWER corrected) more faithfully aligned with grey matter structures after applying the dHCP registration transformations. There was also better separation of activity in physically proximal brain regions that are separated by white matter, which should be devoid of activity e.g. the grey matter of the central sulcus and insular cortex.