This study tested a principle of occupational therapy and motor learning theory in the context of neurodevelopmental treatment techniques. Ten trials of occupationally embedded intervention (playing Simon,™ a computer-controlled game) were compared with 10 trials of rote arm-reach exercise. A counterbalanced design was structured so that each subject experienced each condition one week apart. Subjects were 17 men and 3 women with traumatic brain injury who exhibited mild to moderate spasticity in the upper extremity. Maximum distance from hip to wrist during active reach of the affected extremity was measured by digitization of videotape with the Motion Analysis™ EV-3D system. Results indicated that the use of the game elicited significantly more range of motion than the rote exercise (t (19) = 5.77, p < .001). These results support the use of an occupationally embedded intervention for persons with traumatic brain injury and add to the theoretical base of occupational therapy.