Figure 2.

A, Whole-brain searchlight for cross-format action category decoding. Black arrows and text indicate the anatomical locations of the cross-format clusters identified in this analysis, as well as the location of the ROI for left premotor cortex. Data were corrected for multiple comparisons at p < 0.05 using TFCE and permutation testing. For subsequent analyses, ROIs corresponding to these clusters were defined individually in each subject. B, Whole-brain searchlight for Image Format action category decoding (corrected as in A). C, Whole-brain searchlight for Video Format action category decoding (corrected as in A). D, Conjunction overlap map of the searchlight analyses, with colors indicating which of the three searchlight maps overlap in each area (black outline indicates overlap of all three). Video Format decoding was widespread across the brain, whereas Image Format decoding was mostly restricted to similar regions as were found in the Cross-Format searchlight.