Anatomical and electromyographical evidence suggests a compartmentalized function of the human jaw-closing muscles during both static and dynamic motor tasks. However, the voluntary nature of these tasks hampers unequivocal interpretation of this evidence, because it is impossible to activate voluntarily a single part of a muscle exclusively. Activation of discrete, localized regions can be accomplished with electrical stimulation. A previous study confirmed a functional subdivision of the temporalis muscle into at least three parts. Here, differences in the direction of the lower incisal-point (IP) movement in response to electrical stimulation of four different parts of the masseter muscle were examined in five healthy men. The deep masseter muscle and the anterior, middle, and posterior parts of the superficial masseter muscle were stimulated with monopolar wire electrodes in four different jaw positions (resting position; 50% maximum mouth opening; and 10-mm right and left lateral excursions, both with respect to resting position). Electrode-insertion depth was measured from magnetic resonance images. Movement responses to stimulation were recorded with the OKAS-3D jaw-movement analysis system. The variation in the direction of the IP movement in response to stimulation of parts of the masseter was partly explained by the effects of stimulus location and jaw position. The response to stimulation of the deep masseter was mainly laterovertically directed, whereas the response to stimulation of each of the superficial parts had a mainly anterovertical direction, the responses being most pronounced with the mandible in its resting position. These results provide further evidence for a functional subdivision of the masseter into a superficial part and a deep part, but not for a further subdivision of the superficial part into an anterior, middle, and posterior part.