Generic brain activation mapping in functional magnetic resonance imaging: a nonparametric approach

Abstract

We report a novel method to identify brain regions generically activated by periodic experimental design in functional magnetic resonance imaging data. This involves: 1) registering each of N individual functional magnetic resonance imaging datasets in a standard space; 2) computing the median standardised power of response to the experimental design; 3) testing median standardised power at each voxel against its nonparametrically ascertained distribution under the null hypothesis of no experimental effect; and 4) constructing a generic brain activation map. The method is validated by analysis of 6 null images, acquired under conditions when the null hypothesis was known to be true; 8 images acquired during periodic auditory-verbal stimulation; and 6 images acquired during periodic performance of a covert verbal fluency task.