Brain Research, 188 (1980) 572-577 © Elsevier/North-Holland Biomedical Press
Blindness in monkeys following non-visual cortical lesions
RICHARD K. NAKAMURA and MORTIMER MISHK1N Laboratory of Neuropsyehology, National Institute of Mental Health, Bethesda, Md. 20205 (U.S.A.)
(Accepted December 13th, 1979) Key words: blindness - - vision - - monkey - - cerebral cortex - - attention - - perception - - split-brain
In the course of a study of the visual effects of non-visual lesions in monkeys, we rediscovered a neglected phenomenon that is both highly dramatic and puzzling. As reported initially by Sperry et al. ~5 in cats, and subsequently noted by Gazzaniga 2 in the monkey, large cortical removals outside the modality-specific visual areas can result in a period of total blindness. These authors reached no firm conclusions regarding the basis of the effect and did not investigate it further. In our accidental reproduction of the phenomenon 9, the blindness was found to last from a few weeks in some cases to more than a year in others. During this period the operated animals were completely unresponsive to visual stimuli even though they had an intact geniculostriate pathway and showed normal reactivity to stimuli in other modalities. The results of a preliminary experimental analysis 10 suggest that this phenomenon reflects the loss of an essential nonvisual input to the visual system. We began the study to test whether visual discrimination learning was a localized function in the strictest sense. Earlier research had shown that such learning in the monkey depends not only on striate cortex but also on secondary visual and associated limbic areasS,6, ~6. Other cortical areas, by contrast, seemed unimportant. We wished to know whether the areas that had been identified as necessary for this function were also sufficient, i.e. whether the entire cortical mass outside the necessary areas actually played no appreciable role. To answer the question we removed this cortical mass, sparing only the modality-specific visual areas (striate, prestriate, and inferior temporal) and the limbic areas (cingulate, ventral frontal, and medial temporal). To avoid causing bilateral paralysis, we made this non-visual-non-limbic ablation only in one hemisphere, having first disconnected the other from visual input by severing the optic tract and forebrain commissures. We had anticipated that this combination of lesions would permit the animals to see with one hemisphere and respond with the other1, s. In the initial experiment, two Macaca mulatta (M1 and M2) and two Macaca fascicularis (FI and F2) were trained on a visual pattern discrimination ( + vs E3), after which their right hemispheres were disconnected from vision (see Fig. l a). Following retraining on the discrimination, three of the animals (M1, M2, and F!) received the non-visual-non-limbic ablation in the left hemisphere (see Fig. l a, b). The fourth
Fig. 1. Basic surgical preparation, a: dorsal view. Right hemisphere visually deafferented by right optic tract section and forebrain commissurotomy; interrupted pathways indicated in black. The nonvisual-non-limbic lesion placed in the left hemisphere indicated by stipple, b: lateral and medial views of left hemisphere. The non-visual-non-limbic lesion, shown in stipple, continued to the depth of all sulci, including those at the lesion boundary as indicated by arrows. Abbreviations: AC, anterior commissure; ai, inferior arcuate sulcus; as, superior arcuate sulcus; ca, calcarine fissure; CC, corpus callosum; ce, central sulcus; ci, cingulate sulcus; h, hippocampal sulcus; ip, intraparietal sulcus; 1, lunate sulcus; l a, lateral fissure; oi, inferior occipital sulcus; ot, occipitotemporal sulcus; p, principal sulcus; po, parietooccipital sulcus; rh, rhinal fissure; ts, superior temporal sulcus.
a n i m a l (F2) received the same l e • h e m i s p h e r e a b l a t i o n except t h a t the cortex a n t e r i o r to the a r c u a t e sulcus was spared. W i t h i n two days o f a b l a t i o n all m o n k e y s were able to sit u p r i g h t a n d feed themselves. A t this point, however, it b e c a m e clear t h a t all were blind. W h e n e x a m i n e d in t h e i r h o m e cages, n o n e o f the animals showed a n y detectable reaction to visually presented stimuli, w h e t h e r these were food, fearful objects, threats, or sudden movements. W h e n placed in an u n f a m i l i a r e n v i r o n m e n t , they b u m p e d into obstacles a n d f o u n d scattered p e a n u t s only after t o u c h i n g t h e m accidentally o r exploring for t h e m
574 tactilely with their functionally intact left hands. And when reintroduced to the discrimination apparatus two weeks after the ablation, none could locate more often than by chance which of two uncovered black foodwells contained a white food reward. The only visual response that could be elicited during this period was a weak pupillary reflex. Twenty-five days after surgery one monkey (M l) was able to perform the openfoodwell test at a level just above chance; this was the first indication of vision in any animal. This monkey's ability recovered fully after 18 days of further training, at which point a retest on the pattern discrimination problem revealed good retention. However, this was the only monkey that could be retested with the patterns. The second monkey (M2), having shown no signs of recovery for 42 days, was sacrificed to determine whether there was inadvertent destruction of the primary visual pathway. A third monkey (F2) remained blind for 2l I days and then was also sacrificed for histology. The final monkey (F1) remains alive after more than two years and still cannot perform the open-foodwell test. The duration of the effect makes it unlikely that the blindness is the result of surgical shock 14 or diaschisis3, 7. Nor is the effect due to accidental interruption of the visual pathway. Histological examination in the two monkeys that were sacrificed showed the lesions to be as intended. In one of them (M2) the ventromedial portion of the left lateral geniculate nucleus was partially degenerated indicating some damage to the optic radiation; however, at least 75 ~o of the nucleus remained intact (see Fig. 2a). In the other monkey (F2) the left lateral geniculate was more than 90 ~ intact (see Fig. 2b). Additional evidence that the left geniculostriate pathway was largely preserved in these animals comes from the finding of a clear visual evoked response recorded from the left occiput of the blind monkey (FI) that is still alive (see Fig. 3). Despite evidence of an intact visual pathway, the operated animals' non-reactivity to visual stimuli was indistinguishable from that seen in a control animal prepared like the others with a right optic tract cut and forebrain commissurotomy, but with a complete removal of the striate cortex on the left in place of the non-visual-non-limbic ablation. One difference did emerge, however. The animal that had striate cortex removed tended to stare fixedly, whereas the others moved their heads and eyes freely as if engaged in visual exploration. In this respect their behavior was consistent with the possibility that visual information was being processed normally by an intact visual system in the left hemisphere but could not be communicated to the intact motor system on the right because of the hemispheric disconnection. To test this interpretation we prepared 4 additional monkeys with the same lesions given the first 4, except that we preserved the anterior commissure in one monkey (M3) and all the forebrain commissures in the 3 others (M4, M5, and M6). Since the commissures interconnect both the prestriate and the inferior temporal areas of the two hemispheres 6,12,13, all the animals in the second group had a functional pathway available between the visual system on the left and the motor system on the right. Yet these monkeys were blind like the others for periods of 76 days, 21 days, 7 days, and over one year, respectively. Again the effect occurred despite both electrographic and histologic evidence in each case of a functional visual pathway from the retina to the cortex.
Fig. 2. Coronal sections through left lateral geniculate nuclei, a: monkey M2. Degeneration in medial segment of nucleus from invasion of optic radiations, b: Monkey F2. Cell loss in medial segment only minor.
Fig. 3. Average of 128 evoked responses to light flash recorded over striate cortex in monkey F1. Upper trace: response of striate cortex in left hemisphere associated with blindness produced by nonvisual-nonlimbic lesion. Lower trace: response of striate cortex in right hemisphere associated with blindness produced by optic tract section. Since blindness resulted despite preservation of the forebrain commissures, the effect cannot be due simply to a disconnection of vision from motor mechanisms. The results suggest instead that during blindness the visual signal is not processed beyond the striate cortex, for otherwise the signal should have been transmitted across the preserved commissural channels, which start at the striate-prestriate border. In short, this analysis implies that the blindness reflects a processing failure within the visual system itself, and that the failure occurs no later than prestriate cortex. The most straightforward explanation for such a processing failure in light of both the duration of the blindness and the preservation of the primary visual pathway is that the visual system has been deprived of an essential non-visual input that is normally provided or controlled by the ablated territory. That non-visual cortex should exert some control over processing within the visual system is perhaps not surprising. But that removal of this cortex should have such a drastic effect contradicts traditional views, for it indicates an overriding role of nonvisual cortex in visual perception. If our explanation of the phenomenon is correct, it should be possible to detect the processing failure within the intact visual system of a blind monkey by means of single cell recording. As a preliminary to such an experiment we are attempting to identify the critical lesion for producing the blindness 1° and to obtain preparations in which this effect is reliably long-term 11. Aided by N E I Fellowship 9F32EY-05292 to R.K.N. We thank Dr. Monte Buchsbaum and Dr. Richard Coppola for their assistance in obtaining the visual evoked potentials.
1 Brinkman, J. and Kuypers, H. G. J. M., Cerebral control of contralateral and ipsilateral arm, hand, and finger movements in the split-brain rhesus monkey, Brain, 96 (1973) 653-674. 2 Gazzaniga, M. S., Visuomotor integration in split-brain monkeys with other cerebral lesions, Exp. Neurok, 16 0966) 289-298. 3 Glassman, R. B., Recovery following sensorimotor cortical damage: evoked potentials, brain stimulation and motor control, Exp. Neurol., 33 (1971) 16-29.
577 4 Jones, B. and Mishkin, M., Limbic lesions and the problem of stimulus-reinforcement associations, Exp. NeuroL, 36 (1972) 362-377. 5 Mishkin, M., Visual mechanisms beyond striate cortex. In R. W. Russell (Ed.), Frontiers in Physiological Psychology, Academic Press, New York, 1966, pp. 93-119. 6 Mishkin, M., Cortical visual areas and their interactions. In A. G. Karczmar and J. C. Eccles (Eds.), Brain and Human Behavior, Springer-Verlag, New York, 1972, pp. 187-208. 7 Monakow, C. von, Localization of brain functions. In G. von Bonin (Trans.), The Cerebral Cortex, Charles C. Thomas, Springfield, I11., 1960, pp. 231-250. 8 Myers, R. E., Sperry, R. W. and McCurdy, N. M., Neural mechanisms in visual guidance of limb movement, Arch. Neurol. (Chic.), 7 (1962) 195-202. 9 Nakamura, R. K. and Mishkin, M., Blindness in monkeys after lesions of non-visual cortex, Neurosci..4bstr., 3 (1977) 571. 10 Nakamura, R. K. and Mishkin, M., Blindness in monkeys after lesions of non-visual cortex: not a visual-motor disconnection effect, Neurosci. Abstr., 4 (1978) 639. 11 Nakamura, R. K. and Mishkin, M., Chronic blindness following nonvisual cortical lesions in monkeys, Neurosci. Abstr., 5 (1979) 800. 12 Pandya, D. N., Karol, E. A. and Heilbronn, D., The topographical distribution of interhemispheric projections in the corpus callosum of the rhesus monkey, Brain Research, 32 (1971) 31-43. 13 Rocha-Miranda, C. E., Bender, D. B., Gross, C. G. and Mishkin, M., Visual activation of neurons in inferotemporal cortex depends on striate cortex and forebrain commissures, J. NeurophysioL, 38 (1975) 475-491. 14 Schoenfeld, T. A. and Hamilton, L. W., Secondary changes following lesions: a new paradigm for lesion experimentation, Physiol. Behav., 18 (1977) 951-967. 15 Sperry, R. W., Myers, R. E. and Schrier, A. M., Perceptual capacity of the isolated visual cortex in the cat, Quart. J. exp. Psychol., 12 (1960) 65-71. 16 Sunshine, J. and Mishkin, M., A visual-limbic pathway serving visual associative functions in rhesus monkeys, Fed. Proc., 34 (1975) 440.