To evaluate the effectiveness of a prototype vibrotactile balance prosthesis in maintaining balance during dynamic posturography, we studied 6 subjects with unilateral or bilateral vestibular deficit by means of Equitest computerized dynamic posturography (CDP). Their anterior-posterior (AP) sway at the small of the back was measured with a micromechanical rate gyroscope and a linear accelerometer. The resulting tilt estimate was displayed by a vibrotactile array attached to the torso. The vibration served as tilt feedback to the subject. Subject performance was evaluated with the tilt performance index (TPI), which is the inverse of the root-mean-square of tilt. We found that the balance prosthesis reduced the subjects' AP sway. The subjects' results without the balance prosthesis on CDP sensory organization tests (SOTs) 5 and 6 were compared to results with the prosthesis. The average TPI increased significantly (p < .05) when vibrotactile feedback was used as compared to the unaided condition. This finding was true for both SOTs 5 and 6. We conclude that vibrotactile feedback of estimated AP body tilt improved the subjects' ability to perform selected CDP tests. Some of the subjects were able to stand throughout the test with the device turned on, whereas they otherwise constantly fell.