Effects of occipital lobectomy upon eye movements in primate

1. Eye movements were recorded before and after bilateral occipital lobectomy in six rhesus monkeys trained to fixate and to follow small targets. Striate cortex was completely removed in two animals; small islands remained in the others. In all animals portions of extrastriate cortex were also removed but the medial superior temporal area in the superior temporal sulcus was largely spared. Optokinetic nystagmus (OKN) was markedly altered but not abolished in all animals. The immediate pursuit component of OKN was eliminated leading to a poor response to stimuli comprised of high frequencies. The velocity-storage component of OKN was present, but the maximum value of OKN that could be achieved was decreased to 6 and 16°/s in the most severely affected animals (preop, 65-116°/s). The residual OKN was similar to that of afoveate animals with a diminished response to high velocities of retinal-image motion and a temporal to nasal predominance during mononcular viewing. 2. In the initial postoperative period all animals appeared completely blind. Within 1-6 mo, however, they regained an ability to make visually guided saccades to, and smooth pursuit of, small targets. Saccades were nearly as accurate as preoperatively, but saccade amplitudes were more variable and saccade latencies increased. In the two animals with a complete removal of striate cortex, gains (eye velocity/targaet velocity) of smooth pursuit during sinusoidal tracking (60°/s, 0.5 Hz) were 0.9 and 0.95. During tracking of step-ramp (Rashbass) stimuli with 60°/s ramps, the average acceleration of the eyes during the first 120 ms of smooth pursuit was 189-278°·s^-1·s^-1 (preop range, 154-418°·s^-1·s^-1). In other respects, though, smooth pursuit was not normal. Latencies were increased two- to threefold, and tracking was more variable. 3. Paradoxically, as visually guided saccades and pursuit recovered, some other ocular motor functions deteriorated. Spontaneous and gaze-evoked nystagmus developed 3-6 mo after occipital lobectomy; the time constant of the neural eye-position integrator dropped to values as low as 2.6-4.8 s. The maximum slow-phase velocity of OKN also decreased. 4. The findings immediately after occipital lobectomy indicate that in normal primates occipital cortex is necessary for visually guided saccades and smooth pursuit as well as for the immediate component of OKN. Occipital cortex also makes the predominant contribution toward the generation of the velocity-storage component of OKN. 5. The long-term changes after occipital lobectomy indicate that monkeys can learn to use extrastriate pathways to generate more volitional types of visual-ocular motor behavior such as saccades and smooth pursuit. On the other hand, some of the more reflexive mechanisms that stabilize images on the retina eventually deteriorate, presumably from a chronic loss of visual feedback.