Fig. 2.

VBCM analysis using each speech production measure as the dependent measure [each t-map (scale 3) is thresholded at p < .005, cluster corrected at FWE of p < .001; without lesion volume correction]. Note each result reflects the raw (non-unique) correlation with each speech fluency measure (see text). (A) Lesion correlates for T. (B) Lesion correlates for mean length per utterance, (C) TTR. (D) Lesion correlates for WPM (uncorrected threshold p = .01, voxel-extent 50). (E) Lesion correlates of the unique PCA factors: F1 reflects speech quanta (yellow) and F2 reflects semantic variety (cyan). Note – no significant cluster remains for any factor after lesion volume correction (see text for further details & Fig. 1B).