Figure 4. The positions (column I) and mean parameter estimates of the 12 priori selected ROIs along the left motor (column II) and premotor (column III) strip for the six verb types.

Uncued verbs showed a somatotopic representation among both strips (marked as gray), such that uncued leg verbs showed highest activation in the dorsal regions and uncued mouth verbs showed highest activation the ventral regions. In contrast, an inverse pattern was found for cued verbs, such that cued mouth verbs showed lowest activation in the ventral regions. In addition, ROIs in the middle region showed the strongest overall activation for all six verb types than those in the dorsal and ventral regions. Error bars indicate standard error of the mean. * P < 0.05 ** P < 0.01 *** P < 0.001.