Inhibitory capacity was investigated by measuring the eye movements of normal subjects asked to fixate a central point, and to suppress eye movements toward visual distracters appearing in the periphery or in depth. Eight right-handed young adults performed such a suppression or distracter task. In different conditions, the distracter could appear at 10° left or right at a distance of 20, 40, or 150 cm (calling for horizontal saccades), or in a central position far or close (calling for convergence or divergence), or 7.5° up or down at 40 or 150 cm (calling for vertical saccades). Eye movements were recorded binocularly with an infrared light eye-movement device. Results showed that (1) suppression performance was not perfect, as the subjects still produced eye movements; (2) errors were distributed unequally in three-dimensional space, with more frequent errors toward distracters calling for convergence, or leftward and downward saccades at a close distance; (3) distracters calling for saccade suppression yielded saccades in the direction of the distracter (that we called prosaccades), and saccades directed away from it (that we called spontaneous antisaccades); (4) for vergence, only distracters calling for convergence yielded errors, which were always promovements; (5) in addition, a small convergent drift was found for convergence distracters. Differences in the errors between saccade and vergence suggest that different inhibitory mechanisms may be involved in the two systems. Spatial left/right, up/down, and close/far asymmetries are interpreted in terms of attentional biases.