Two major concerns with respect to task-based motor functional magnetic resonance imaging (fMRI) are inadequate participants' performance as well as intra- and inter-subject variability in execution of the motor action.This study validates the use of an MRI-compatible stimulator based on a pneumatic artificial muscle (PAM) for block-design fMRI mapping of the primary sensorimotor (SM1) cortex in a series of fifteen right-handed healthy subjects. The PAM stimulator elicits computer-controlled timely and reproducible passive movements of fingers/toes. Participants performed comparable active and passive PAM-induced flexion-extensions of the index fingers.Passive movement of the right index finger and passive alternating right and left index finger movement resulted in a significant increase in blood-oxygen-level-dependent (BOLD) signal in contralateral SM1 cortex in 14/15 and 15/15 subjects respectively. Similar networks were recruited by active and passive index finger movements. However, at the group level, active movement induced significantly higher increases in BOLD signal than passive movement in contralateral SM1 cortex (p < 0.05 Family Wise Error [FWE] corrected), supplementary motor area (p < 0.001 uncorrected), ipsilateral cerebellum (p < 0.001 uncorrected), and bilateral putamina (p < 0.001 uncorrected).As compared to the several MRI-compatible robotic devices for computer-controlled passive movement of the fingers that were introduced in the past decades, the proposed PAM-based stimulator is smaller, handier, and easier to use in the MRI setting.PAM-based stimulators can be reliably used for passive sensorimotor fMRI mapping in healthy subjects. Using this approach, bilateral SM1 cortices can be mapped accurately during a single 6-min block-design fMRI protocol.