Three groups of ten subjects each performed a unidimensional compensatory tracking task for 30 minutes. Each group received continuous error information feedback either visually, auditorally, or cutaneously. Direction of error was signaled by the side of the body stimulated, magnitude of error was indicated by signal intensity. Intensities of the signals were subjectively matched. Performance was measured by time on target, constant error, absolute error, and root-mean-square error. Analyses of variance for these measures yielded significant differences between display types on all but the root-mean-square measure. A multiple discriminate analysis was performed on these measures and provided two significantly independent dimensions of discrimination. The first dimension was defined as “speed of response,” the second “goodness of performance.” It was concluded that for the particular task used, the auditory and cutaneous displays proved more effective than the visual display.