Skip to main content
. Author manuscript; available in PMC: 2015 Apr 3.
Published in final edited form as: Neurosci Lett. 2014 Feb 7;564:21–26. doi: 10.1016/j.neulet.2014.01.056

Figure 1.

Figure 1

Overview of methodological approach. (Blue arrows) Task-based recruitment of an ICA component is traditionally assessed by first back-projecting the ICA spatial map (via multiplication with the nth ICA component’s spatial map) to each timepoint in the fMRI timeseries, thus generating a 1D timeseries weighted by the component. Univariate GLM then determines an activity beta-value (activity βs) and t-score describing that component’s recruitment by one or more task contrasts. (Red arrows) We propose an alternative approach whereby the fMRI timeseries undergoes whole-brain GLM to generate a spatial map for each GLM contrast. The ICA component is then back-projected (again via multiplication) to produce activity βs for that component. Abbreviations: ICA, independent component analysis; fMRI, functional MRI; 1D, one-dimensional; GLM, general linear modeling.