Comparing performance of different subspace estimation methods, for Strong-Contrast data.

<p>For <i>strong</i> contrast (Task vs. Control), model performance is shown for different subspace estimation methods, relative to full-dimensionality data (i.e. retaining all PCs). The median, [minimum, maximum] changes are shown for prediction (<b>Δ</b><i>P</i>), reproducibility (<b>Δ</b><i>R</i>) and distance <b>Δ</b><i>D</i> from (<i>P</i> = 1,<i>R</i> = 1), over all single-subject results. Significance is given by Wilcoxon tests, with * indicating significant improvement. We show results for combinations of <b>ICA</b> = MELODIC subspace estimation, <b>PCA_split</b> = optimized PC subspace on each data split-half, and <b>PCA_full</b> = retaining 35% of PCs from the full data matrix. Note that (<b>PCA_full</b>+<b>PCA_split</b>) is the subspace selection method used for the rest of the manuscript. Results are shown for optimal fixed preprocessing: motion correction and 2^nd-order detrending.