J. Physiol. (1976), 257, pp. 155-170 155 With 5 text-figures Printed in Great Britain KITTENS REARED IN A UNIDIRECTIONAL ENVIRONMENT: EVIDENCE FOR A CRITICAL PERIOD BY N. W. DAW AND H. J. WYATT* From the Department of Physiology and Biophysics, Washington University Medical School, St Louis, Missouri 63110, U.S.A. (Received 16 September 1975) SUMMARY 1. Kittens were reared in the dark from birth except for a period each day when they were put inside a stationary transparent cylinder, around which a drum, with vertical black and white stripes on the inside, rotated in one direction. After the end of the period of exposure, we recorded a sample of single cells from their visual cortices, and analysed each cell for direction and orientation sensitivity and other properties. 2. Two kittens were placed inside the drum, rotating rightward, for 2 hr each weekday from 3j to 7 weeks of age. A greater proportion of the directionally sensitive cells in their cortices showed a preference for rightward movement. 3. Six other kittens were placed inside the drum for 1 hr each weekday from 2 to 12 weeks of age with the drum rotating leftward up to a particular changeover age, then rightward until 12 weeks. The changeover point occurred at 21, 26, 28, 33, 35 and 51 days for different kittens. A changeover earlier than 4 weeks of age led to a preponderance of cells preferring rightward movement. A changeover later than 5 weeks of age led to a preponderance of cells preferring leftward movement. Comparison of these results with others on monocular deprivation suggests that the peak of the critical period for directional deprivation may occur earlier than the peak of the critical period for monocular deprivation. 4. None of the samples of cells showed a preponderance of cells specific for vertical orientations. It is unclear whether this negative effect resulted from the presence of some horizontal contours during exposure, or some more fundamental cause. * Present address: State University of New York, State College of Optometry, New York, New York 10010.
156 156 ~N. W. DAW AND H. J. WYATT INTRODUCTION Several different kinds of visual deprivation lead to physiological changes in the visual cortex of the kitten. After monocular deprivation most cells in the cortex cannot be driven through the deprived eye and the animals are almost blind in that eye (Wiesel & Hubel, 1963; Dews & Wiesel, 1970). Raising kittens in a striped environment leads to most cells having preferred orientations the same as the stripes, and only a few having preferred orientations orthogonal to this (Hirsch & Spinelli, 1970, 1971; Blakemore & Cooper, 1970). Raising in an environment moving unidirectionally leads to the majority of cells preferring movement in the direction of exposure (Tretter, Cynader & Singer, 1975; Cynader, Berman & Hemn, 1975). Other types of deprivation also have specific effects (Shlaer, 1971; Cynader, Berman & Hein, 1973; Freeman & Pettigrew, 1973; Pettigrew & Freeman, 1973; Van Sluyters & Blakemore, 1973; Olson & Pettigrew, 1974). There is a critical period for the effects of monocular deprivation. The cortex is most susceptible between the fourth and sixth weeks of age, and after the twelfth week little effect is seen (Hubel & Wiesel, 1970; Blakemore & Van Sluyters, 1974; Olson & Freeman, 1975). Around the fourth week, other kinds of deprivation are effective when applied for short periods of only a few hours (Blakemore & Mitchell, 1973; Blakemore, 1974; Tretter et al. 1975). However, no close comparison of the critical periods for monocular deprivation and other kinds of deprivation has been reported. Our experiments confirm that rearing in a unidirectional environment is an effective deprivation and suggest that the critical periods for monocular and directional deprivation may not be the same. These experiments were started while we were rearing rabbits in a unidirectional environment to see if there was any effect on the directionally selective cells of the rabbit retina (Daw & Wyatt, 1974). When it became apparent that the effect on rabbit retina would probably be negative we were naturally anxious to see if a positive effect would be obtained in the cat cortex. Both sets of experiments were done with the same apparatus. While differences in receptive field and eye movement properties of the two species make the experiments not strictly comparable, the results are definitely positive for the cat cortex and negative for the rabbit retina. METHODS Animal&. The mother was moved into a dark room shortly after delivery of the kittens and stayed there with them until weaning time. The dark room was inside a larger room where the lights could be turned off, so that the animals could be fed and cages cleaned without exposure to light.
VISUAL DEPRI VATION 157 Selective visual experience. The drum (Daw & Wyatt, 1974) contained vertical black stripes on a white background, subtending between 1 9 and 180 at the kitten's eye, and extending from 20 below to 650 above the level of the kitten's eye. The kitten was placed on a platform in a transparent cylinder of 9 in. diameter inside the drum, which was 20 in. diameter. The kitten was prevented from seeing out of the drum upwards or downwards by the roof and platform, both of which were black. However an observer could see the kitten's limbs, and observe if it was sitting upright or lying on its side. An observer was in the room all the time when the kittens were in the drums, to take immediate action if there were problems, and looked at them every 10 or 15 min. Experiments on a test litter of kittens showed that they do not have sustained optokinetic nystagmnus, and do not follow the drum at high velocities, as rabbits do (Daw & Wyatt, 1974). A similar result for kittens has been noted by Vital-Durand & Jeannerod (1974a): optokinetic nystagmus would be there for a few min at the start of each session, after which it would be intermittent. A speed of rotation of 30'/sec was found to be sufficiently fast that the drum rotated faster than the eyes, and is still within the range of response of most complex cortical cells (Movshon, 1975). During the periods of optokinetic nystagmus, the velocity of stripes moving across the retina was much less than 30'/sec, and would have stimulated the simple cells as well. We also found, with the test litter of kittens, that head restraining devices irritated the kittens considerably. They would twist their heads around, and sometimes end up upside down. Consequently, we fitted them with ruffs (Blakemore & Cooper, 1970) which not only prevented them from observing their limbs, but also made lying down sufficiently uncomfortable so that they always sat upright. Recording. The cat was anaesthetized with halothane, and paralysed with D-tubocurare (3 mg/hr) after a tracheal cannula was inserted. Anaesthesia was maintained with 70 % N20 and 30 % 02 administered by a respiration pump. Wound areas were infiltrated with a long lasting procaine anaesthetic. Temperature was maintained with a heating pad. Tracheal CO2 concentration was checked periodically with an infra-red CO2 analyser (Beckman Instruments), and kept at 4-5 % by adjusting the rate of respiration or, in some cases, changing to 70 % N20, 28-5 % 02' 1.5 % C02. The animal was placed in a stereotaxic instrument. After removing bone and dura, a transparent chamber was cemented to the skull and filled with warm mineral oil. The electrode was inserted through a ball joint in the chamber, and angled to go down the medial bank of the lateral gyrus. Most penetrations were made between Al and P2 into the left cortex. Electrodes were lacquered tungsten (Hubel, 1957), connected to conventional amplification and display equipment. Analysis of receptive fields. The kitten's eyes were focused on a tangent screen about 40 in. away with contact lenses, the correct power being determined by streak retinoscopy. Pupils were dilated with atropine and nictitating membranes withdrawn by neosynephrine. Receptive fields were plotted by hand on the tangent screen after moving spots and bars of various sizes across the screen in various directions at various velocities. For directionally sensitive units, the preferred direction was determined by moving a long bar through the receptive field perpendicular to its length in various directions, finding those directions that gave a just noticeable response, and bisecting the angle between. (If a unit only responded to bars shorter than a particular length, the maximum length possible was used.) For orientation sensitive units, the preferred orientation was measured in a similar manner, flashing a stationary bar at various orientations, finding those orientations that gave a just noticeable response, and bisecting the angle between. Sometimes the thresholds were indistinct, in which case the preferred orientation was taken to be perpendicular to the preferred directionss.
N. W. DAW AND H. J. WYATT Histology. After recording, each cat was perfused with formol saline, and the visual 158 cortex was sectioned and stained to determine whether the electrode had been in area 17 or area 18. Lesions (5,sA for 5 sec) marked points along some electrode tracks, to indicate the depth of particular points. I 200 jlm Area 17 Fig. 1. Penetration from a cat with leftward, followed by rightward, ex perience, showing the spacing of the cells recorded. Each single arrow represents a unidirectional cell; each double arrow represents a bidirectional cell; a bar without arrows on it represents the preferred orientation for a stationary flashed bar. Precautions against sample and experimenter bias. The second and third litters were recorded without the experimenter knowing which animal was being recorded. A technician would bring the kitten up from the dark room, and not tell the experimenters which animal was which until recordings had been made from the complete litter. Long penetrations were made down the medial bank of the lateral gyrus, so that as many cortical columns as possible would be crossed. After each cell was analysed, the electrode was advanced 150 ysm before isolating another cell (Fig. 1). However, cells responding to both possible directions are found within a single
VISUAL DEPRIVATION orientation column, for example the first cell in the penetration of Fig. 1 responded to movement left and the fourth responded to movement right, both within a column specific for vertical orientations. Consequently, the possible bias introduced into the results by recording several units from within a single column is much less for direction sensitivity than it is for orientation sensitivity. RESULTS Classificaion of cells Almost every author who has worked on the cat visual cortex has a different set of criteria for classifying cells. Our main aim was to record enough cells in each animal so that the results would be significant. Consequently we decided to adopt criteria that would allow us to classify cells in a reasonably short period of time. Directional sensitivity. We divided cells into unidirectional, bidirectional and non-directional. Unidirectional cells were those which gave a response for movement in the preferred direction, and little or no response for movement at 1800 to this (the null direction). This corresponds quite closely to the criterion for directional selectivity used by Pettigrew, Nikara & Bishop (1968) and Bishop, Coombs & Henry (1971a). Bidirectional cells were those which responded to movement of a long bar moving in a particular direction or its reverse, but not at 900 to this. Non-directional cells did not show a marked preference for movement of a spot or bar in any particular direction. Orientation sensitivity. Cells can be analysed for orientation sensitivity with stationary flashed bars, or moving bars. When moving stimuli are used the cell is usually said to be orientation sensitive if the response to a bar is greater than the response to a spot, if the response is bidirectional, or if the tuning for directional selectivity is tighter for a bar than for a spot (Hubel & Wiesel, 1962, 1965; Pettigrew et al. 1968; Blakemore & Van Sluyters, 1974; Cynader et al. 1973, 1975). Using moving bar criteria, nearly all cells in the cortex are orientation sensitive. The stationary bar tests cannot be applied to those complex cells that do not respond to stationary bars. For some cells, like rabbit directionally sensitive ganglion cells, the criteria conflict: there is a response for all orientations of a stationary stimulus, but the tuning for a moving bar is narrower than the tuning for a moving spot (Wyatt & Daw, 1975). How many of the cells in the cat cortex fall into this category has not yet been determined. We measured the preferred orientation of cells using stationary stimuli, since it can be done most rapidly by this means. The distributions of preferred orientations which follow reflect this. However, none of the conclusions are altered if cells classified as orientation sensitive by moving bar criteria are included, with the assumption that the preferred orientation 6 PHY 2,57 159
160 N. W. DAW AND H. J. WYATT is perpendicular to the preferred direction. It should be emphasized that nearly all cells with a preferred orientation were also either unidirectional or bidirectional (Table 1). Other properties. Cells were also tested for their ocular dominance and complexity (simple, complex or hypercomplex) according to the criteria of Hubel & Wiesel (1962, 1965), their spontaneous activity, their preferred speed of movement, and the range of velocities over which they would respond. No significant trends were noticed when the data were split up and studied by these criteria. TABLE 1. Numbers and percentages of cells for total sample, normal and deprived, classified according to their directional properties for a moving bar, and orientation properties for a stationary bar. For stationary bars of light Orientation sensitive Non-oriented Unresponsive Unclassified Total Unidirectional 111 (29%) 12 (3 %) 54 (14%) 46 (12%) 223 (58%) Bidirectional 91 (24%) 8 (2%) 9 (2%) 26 (7 %) 134 (35%) Non-directional 2 (1%) 20 (5%) 2 (1%) 1 (0%) 25 (7 %) Total 204 (53%) 40 (10%) 65 (17%) 73 (19%) 382 Total sample of cells. The percentage of cells in each direction and orientation category did not vary significantly between the normal and the various categories of deprived animals (see below) and the statistics for the complete sample is given in Table 1. The percentage of unidirectional cells (69 % in recordings from ten normal animals, 56-58 % in deprived ones) compares with that from Pettigrew et al. (1968) which was 42% total and that from Bishop et al. (1971 a) which was 72 % for simple cells. It is less than the percentage observed by Cynader, Berman & Hein (1973) who found 83 % using a broader criterion. Bias of preferred directions of unidirectional cells in normal animals. Nearly all our recordings were made in the left cortex. One might well wonder if there is a bias among these cells for movement right, since the left cortex projects to the left colliculus where such a bias is found (Sterling & Wickelgren, 1969; Berman & Cynader, 1972). Our results indicate that there probably is such a bias (Fig. 2). Excluding cells with directions within 150 of the vertical, 60 % (26/43) prefer movement right and 40 % (17/43) prefer movement left. Palmer & Rosenquist (1974) studied cortical cells from the left cortex projecting to the superior colliculus, and in their sample 64 % (18/28) preferred movement right and 36 % (10/28) preferred movement left (Fig. 2), although they did not feel that the bias was strong enough to make a point about it. The
VISUAL DEPRIVATION 161 corticotectal cells are a small proportion of the total number of cells in the cortex, but may be a rather larger proportion of the unidirectional ones. Animals raised in a unidirectional environment T'wo kittens were placed in the drum from 31 to 7 weeks of age for 2 hr each weekday, with the drum moving rightwards at 30'/sec. They were then kept in the dark for a period of time, and recordings were made from their cortices at the age of 9 months. A B 17 =:\\=tz 26 10 Fig. 2. Preferred directions of directionally selective cells in left cortex of normal cats. A, our data; B, data re-drawn from Palmer & Rosenquist (1974), Fig. 5. Each arrow represents the preferred direction of a cell with unidirectional properties. Numbers give the totals preferring leftward or rightward movement, excluding those within 15 of the vertical (covered with a screen). Before recording, a few gross behavioural observations were made. Little placing reaction was evident. They would bump into objects as they moved around the room, and start to walk off the side of the stool when placed on top of it. These deficits are common to all visually deprived animals, whether monocularly (tested with the deprived eye), binocularly, orientationally, or directionally deprived (Wiesel & Hubel, 1963, 1965; Blakemore & Cooper, 1970). Both animals were tested for following movements of the head and eyes. A piece of black paper, 2 in. x 5 in., on the end of a yardstick was moved across the field of vision from left to right and from right to left in a random sequence. They followed it more frequently from left to right than from right to left. This tallies with observations on optokinetic nystagmus in unidirectionally reared animals (Vital-Durand & Jeannerod, 1974b). Results from the cortical recordings in these animals show that nearly 6-2
162 N. W. DAW AND H. J. WYATT all the unidirectional cells preferred movement to the right (Fig. 3). These recordings were made in the left cortex, where there is a tendency for unidirectional cells in normal animals to prefer rightward movement (Fig. 2), but the tendency was considerably more marked in the deprived animals. Orientation Unidirectional Bidirectional sensitive cells cells cells Normal cats l Rightward I reared ----- cats Fig. 3. Preferred directions and orientations of cells from the left cortex of normal cats, and two cats reared in a unidirectional environment. Each unidirectional cell represented by a single arrow pointing in the preferred direction. Each bidirectional cell represented by a double headed arrow similarly. Each orientation sensitive cell represented by a line along the axis of best response for a flashed bar. The percentage of unidirectional and bidirectional cells does not appear to have been altered significantly by the deprivation (Table 2). In both cases most of the cells were unidirectional, and about a quarter bidirectional. The group of non-directional cells in the rightward-raised animals may have been inflated by inclusion of some lateral geniculate afferents Surprisingly, the distribution of preferred orientations in the drumreared animals did not show a bias towards vertical orientations (Fig. 3), even though the drum had only vertical stripes on it. This puzzling result is discussed below.
VISUAL DEPRIVATION 163 Animals raised in a leftward drum followed by a rightward drum Six other kittens (two litters of three kittens each) were reared with experience in a drum moving left for a period of weeks, followed by a drum moving right for a period of weeks. Each animal was taken from the dark and put into the drum for the first session at the age of 14 days. Each session lasted 1 hr. and each animal was given one session per day, Monday to Friday. The direction of rotation of the drum was left for all animals for the initial sessions. For one animal, the direction of rotation of the drum was reversed after the twenty-first day of age, for the others after the twenty-sixth, twenty-eighth, thirty-third, thirty-fifth or fiftyfirst days of age. Daily sessions were then continued until the animals were 12-13 weeks old (85 days for the first litter, 91 days for the second). The animals were then kept in the dark until recording, at the age of 4-7 months. TABLE 2. Percentages of the three categories of directional sensitivity found in normal animals and the two groups of experimental animals Animals reared in Normal animals Animals reared in rightward drum rightward drum then leftward drum Unidirectional 64 (69%) 30 (58%) 129 (56%) Bidirectional 23 (25%) 13 (23%) 98 (42%) Non-directional 6 (6%) 14 (24%) 5 (2%) Total 93 57 232 Where reversal occurred before the fifth week, the majority of unidirectional cells responded to movement right (Fig. 4). Where reversal occurred after the fifth week, the majority of these cells responded to movement left, even in cases (animal E for example) where the animal spent more time in a rightward moving drum than a leftward moving drum. Consequently exposure around the fourth week appears to be particularly important for the final determination of the preferred directions of the unidirectional cells. Again, the percentages of the various types of direction sensitive cells is not significantly different from normal (Table 2). Also, surprisingly again, there was no definite tendency towards preference for vertical among the orientation selective cells (Fig. 4). Only the sample in animal D showed a marked tendency towards a vertical preference; the tendency in the samples from animals A, C and F was slight, and in animal E the sample tended towards horizontal preferences. These results can be plotted to give an indication of the peak of the
164 164 ~N. W. DAW AND H. J. WYATT Unidirectional cells Bidirectional cells Orientation sensitive cells O 14 21 85 Left 0 14 26 91 x C V//4%I RightJ 0 14 28 91 0 14 33 91 ;" Left I Right. I + -, F Left Age (days) Right-.&g Fig. 4. Preferred directions and preferred orientations of cells in the cortex of kittens reared undirectionally with reversal of direction at various ages. Direction of drum rotation and timing in days given on left.
VISUAL DEPRIVATION 165 critical period, by comparing the number of unidirectional cells preferring rightward movement with the number preferring leftward movement in each animal. Firstly, one excludes those cells with preferred directions within 150 of the vertical on the grounds that: the kittens did not hold their heads precisely upright while in the drum; there is an uncertainty of 50 or 100 in determining the preferred direction; the eyes may rotate a few degrees when paralysed. Then the number of units in the rightward group (expressed as percentage of the number of units in rightward plus leftward groups) is plotted against the time at which reversal of drum rotation occurred (Fig. 5). Also plotted in Fig. 5 is the reversal index from reversals of monocular deprivation measured by Blakemore & van Sluyters (1974). Although the results are not strictly comparable (see Discussion), they do suggest that the critical period for directional deprivation may be earlier than the critical period for monocular deprivation. 100 80-0 D Do EU,,40_()@ _\_ XL 20 \ i< _ 20 0 IIIII 0 2 t4 6 Jf 8 10 12 14 Age at reversal (weeks) Fig. 5. Plot of percentage of cells preferring rightward movement v8. age at which drum rotation was reversed from left to right (letters and continuous curve). Letters identify individual animals from Fig. 4. Also shown is the reversal index for monocular reversals from Blakemore & van Sluyters (1974), Fig. 3 (circles and dashed curve). The horizontal lines show the percentages for normal animals (continuous line, percentage of cells preferring rightward movement; dashed line, percentage of cells dominated by the right eye). Arrows indicate reversal times giving rise to normal distributions of preferred directions or ocular dominance. DISCUSSION These results show that after kittens are exposed to stripes moving right for a period of time each day, this being their only exposure to light, most of the cells in their cortex which have a strong preference for movement in one direction prefer movement right rather than left. This agrees with the
166 N. W. DAW AND H. J. WYATT results of Tretter et al. (1975) and Cynader et al. (1975). The parameters of the rearing conditions in the three studies were somewhat different. Tretter et al. gave exposures of 3 hr per day for 1, 2 or 4 days at around 4 weeks of age. Their drum moved at 150/sec, and had black and white stripes of equal width, 50 each. Cynader et al. (1975) used an enclosure which was dark, except for a pattern of luminescent irregularly-shaped patches. The kittens were free to move around and kept in the enclosure almost continuously. Depending on the position of the kitten, the velocity was 9-360/sec. This compares with our pattern of vertical lines of various widths (1.9-18o) moving at 300/sec. Thus the result is obtained in a wide variety of conditions. Our positive results were obtained in the same apparatus which gave negative results for directionally sensitive ganglion cells in the rabbit retina (Daw & Wyatt, 1974). The circumstances were slightly different: the velocity of drum rotation was faster for the rabbit (up to about 1000/sec) and since rabbits tended to follow the drum with a nystagmus that had three components, following, stationary and flick-back, the velocity of movement across the retina was small for part of the time while following, 100'/sec for part while stationary and about 2000/sec during the flick-back. The kittens exhibited intermittent nystagmus after an initial period, so that the velocity across their retina was 30'/sec for most of the time, interspersed with some periods with a much lower velocity. In both situations, however, the velocity across the retina was low enough to stimulate the cells involved (rabbit retinal directionally sensitive cells, at least the on-off type, respond to higher velocities of movement than most cat area 17 directionally sensitive cells). Orientation sensitivity. A puzzling finding in the present experiments is that the sample of orientation sensitive cells did not show any tendency towards vertical orientations. The data shown are for cells classified as orientation sensitive using stationary bars of light, but if the broader class of cells is analysed, including those classified as orientation sensitive using moving bars of light, the conclusion is the same. Tretter et al. (1975) found a positive bias towards vertical orientation in their experiments, and so did Cynader et al. (1975), even though the stimulus in the latter case was a series of blobs rather than vertical stripes. One possible explanation is that our kittens saw some horizontal lines during their period of rearing, and this was enough to prevent the synaptic connexions for horizontal orientations from decaying. The base and the roof of the cylinder in which the kittens sat were both black, with horizontal edges that the kittens could see. The edge of the cuff was horizontal above and below, although it was hard for the kitten to look at it with the centre of its field of vision, where most of the receptive fields that we
VISUAL DEPRIVATION 167 recorded were situated. Perhaps the roof and floor provided adequate exposure. Positive results from orientation deprivation have recently been questioned by Stryker & Sherk (1975) who were careful to avoid experimenter and sampling bias. They reared four kittens in a horizontal environment, three in a vertical environment, recorded 456 units in them all, and found no over-all tendency towards orientation preference along the axis of the rearing environment. This must be put against several studies which have found positive results in cortical recordings from kittens raised in a striped environment (Blakemore & Cooper, 1970; Hirsch & Spinelli, 1970, 1971; Blakemore & Mitchell, 1973; Pettigrew, Olson & Hirsch, 1973; Blakemore & Papaiannou, 1974; Tretter et al. 1975; Blakemore & Van Sluyters, 1975), and one study which showed that such kittens have a small but significant and permanent reduction in acuity for gratings oriented along the axis of deprivation (Muir & Mitchell, 1975). In designing our conditions of rearing, we were primarily concerned with directional deprivation and ensuring that wrong-way movement of the stimulus across the retina did not occur. We do not think that our animals were reared carefully enough, as far as orientation deprivation is concerned, to provide any solid evidence with regard to that controversy. Irrespective of the final resolution of the orientation deprivation controversy, it is interesting that it is possible to rear a litter of kittens with direction sensitivity altered and orientation sensitivity normal. It is also interesting that the percentage of unidirectional vs. bidirectional cells was not altered. Perhaps only unidirectional cells were altered by our procedures, and the alteration was limited to unidirectional cells with a leftward preference being converted into unidirectional cells with a rightward preference, without any change in the orientation sensitivity of any of the cells involved. In this case, of course, the distribution of preferred orientations would have to be unchanged. The interpretation of these results in terms of cell connexions will have to wait for more fundamental work. While there is a model for orientation sensitivity in terms of simple, complex and hypercomplex cells (Hubel & Wiesel, 1962, 1965) and a model for directional sensitivity (Bishop, Coombs & Henry, 1971 b), both of these models rest on circumstantial rather than direct evidence. Speculation of which synapses might be altered should wait for more evidence on which synapses are involved in the normal animal. Critical period. There is a suggestion, from the comparison between our results on unidirectionally deprived animals with a reverse of direction and Blakemore & Van Sluyters' (1974) results on monocularly deprived animals with reverse suture, that the critical period for directional
168 N. W. DAW AND H. J. WYATT deprivation may occur earlier than the critical period for monocular deprivation. One might expect that critical periods for different types of deprivation might occur at different times, if different properties of cortical cells develop at different times (Pettigrew, 1974). The comparison between the Blakemore & Van Sluyters' curve and ours is not precise, because each of their animals was exposed for 9 weeks after reverse suture. However, if one compares the animals reversed at 5 weeks, where the time of exposure after reversal was approximately the same, there is a significant difference between monocularly and unidirectionally reared animals. One also has to take into account the slight tendency in normal animals for cells to be driven by the contralateral eye, and prefer movement in the contralateral direction. Nearly all our recordings were from the left cortex, where, in normal animals, about 60 % of the unidirectional cells prefer movement to the right (see Results). The Blakemore & van Sluyters' recordings were from the right cortex, where only 37 % of the cells (Hubel & Wiesel, 1962, excluding cells in group 4), are normally dominated by the right eye. Consequently one should compare the position at which our curve crosses the 60 % line with the position at which their curve crosses the 37 % line. This accentuates the difference between the two sets of results. Several factors might account for the difference between the two types of deprivation. There can be differences in the critical period between litters (Hubel & Wiesel, 1970). The time of exposure per day, or the physically confining circumstances of the directional deprivation might affect the results. Alternatively, there may be a real difference in the cortical response to monocular and directional deprivation. Experiments are now under way to make a comparison in which all variables except the last one are, as far as possible, equal. We thank Eileen Roach and Marcy Holm for help in the rearing of kittens, and doing histology on their brains. Dr Nancy Berman and Dr Colin Blakemore provided helpful comments on the manuscript. The work was supported by an NIH Research Grant EY00053 and Training Grant NS 05644. REFERENCES BERMAN, N. & CYNADER, M. (1972). Comparison of receptive field organisation of the superior colliculus in Siamese and normal cats. J. Phy~iol. 224, 363-389. BISHOP, P. O., Coomis, J. S. & HENRY, G. H. (1971a). Responses to visual contours: spatiotemporal aspects of excitation in the receptive fields of simple striate neurones. J. Phy8iol. 219, 625-657. BISHOP, P. O., COOMBS, J. S. &EHENPY, G. H. (1971 b). Interaction effects of visual contours on the discharge frequency of simple striate neurones. J. Phy8iol. 219, 659-687.
VISUAL DEPRIVATION 169 BLAKEMORE, C. B. (1974). Developmental factors in the formation of feature extracting neurons. In Neurosciences: A Third Study Program, ed. SCHMIDT, F. O. pp. 105-113. Cambridge, Mass.: M.I.T. Press. BLAKEMORE, C. B. & COOPER, G. F. (1970). Development of the brain depends on the visual environment. Nature, Lond. 228, 477-478. BLAKE MORE, C. B. & MITCHELL, D. E. (1973). Environmental modification of the visual cortex and the neural basis of learning and memory. Nature, Lond. 241, 467-468. BLAKEMORE, C. & PAPAIANNOU, J. (1974). Does the vestibular apparatus play a role in the development of the visual system? J. Phy8iol. 236, 373-385. BLAKEMORE, C. & VAN SLUYTERS, R. C. (1974). Reversal of the physiological effects of monocular deprivation in kittens; further evidence for a sensitive period. J. Physiol. 237, 195-216. BLAKEMORE, C. & VAN SLUYTERS, R. C. (1975). Innate and environmental factors in the development of the kitten's visual cortex. J. Phyriol. 248, 663-716. CYNADER, M., BERMAN, N. & HEIN, A. (1973). Cats reared in stroboscopic illumination: effects on receptive fields in visual cortex. Proc. natn. Acad. Sci. U.S.A. 70, 1353-1354. CYNADER, M., BERMAN, N. & HEIN, A. (1975). Cats raised in a one-directional world: effects on receptive fields in visual cortex and superior colliculus. Expl Brain Re8. 22, 267-280. DAW, N. W. & WYATT, H. J. (1974). Raising rabbits in a moving visual environment: an attempt to modify directional sensitivity in the retina. J. Phy8iol. 240, 309-330. DEWS, P. B. & WIESEL, T. N. (1970). Consequences of Monocular deprivation on visual behaviour in kittens. J. Physiol. 206, 437-455. FREEMAN, R. D. & PETTIGREW, J. D. (1973). Alteration of visual cortex from environmental asymmetries. Nature, Lond. 246, 359-360. HIRSCH, H. V. B. & SPINELLI, D. N. (1970). Visual experience modifies distribution of horizontally and vertically oriented receptive fields in cats. Science, N.Y. 168, 869-871. HIRSCH, H. V. B. & SPINELLI, D. N. (1971). Modification of the distribution of receptive field orientation in cats by selective visual exposure during development. Expi Brain Res. 13, 509-527. HUBEL, D. H. (1957). Tungsten microelectrode for recording from single units. Science, N.Y. 125, 549-550. HUBEL, D. H. & WIESEL, T. N. (1962). Receptive fields, binocular interaction, and functional architecture in the cat's visual cortex. J. Physiol. 160, 106-154. HUBEL, D. H. & WIESEL, T. N. (1963). Receptive fields of cells in striate cortex of very young, visually inexperienced kittens. J. Neurophysiol. 26, 994-1002. HUBEL, D. H. & WIEFEL, T. N. (1965). Receptive fields and functional architecture in two non-striate visual areas (18 and 19) of the cat. J. Neurophy8iol. 28, 229-289. HUBEL, D. H. & WIESEL, T. N. (1970). The period of susceptibility to the physiological effects of unilateral eye closure in kittens. J. Physiol. 206, 419-430. MOVSHON, J. A. (1975). The velocity turning of single units in cat striate cortex. J. Phy&iol. 249, 445-468. MUIR, D. W. & MITCHELL, D. E. (1975). Behavioural deficits in cats following early selected visual exposure to contours of a single orientation. Brain Res. 85, 459-477. OLSON, C. R. & FREEMAN, R. D. (1975). Progressive changes in kitten striate cortex during monocular vision. J. Neurophysiol 38, 26-32. OLSON, C. R. & PETTIGREW, J. D. (1974). Single units in visual cortex of kittens reared in stroboscopic illumination. Brain Res. 70, 109-204.
170 N. W. DAW AND H. J. WYATT PATER, L. A. & ROSENQUIST, A. C. (1974). Visual receptive fields of single striate cortical units projecting to the superior colliculus in the cat. Brain Re8. 67, 27-42. PETTIGREW, J. D. (1974). The effect of visual experience on the development of stimulus specificity by kitten cortical neurones. J. Physiol. 237, 44-74. PETTIGREW, J. D. & FREEMAN, R. D. (1973). Visual experience without lines: effect on developing cortical neurons. Science, N.Y. 182, 599-601. PETTIGREW, J. D., NiKARA, T. & BIsHOP, P. 0. (1968). Responses to moving slits by single units in cat striate cortex. Expl Brain Re8. 6, 373-390. PETTIGREW, J. D., OLSON, C. & HIRsCH, H. V. B. (1973). Cortical effect of selective visual experience: degeneration or reorganization? Brain Re8. 51, 345-351. SMrIR, R. (1971). Shift in binocular disparity causes compensatory change in the cortical structure of kittens. Science, N.Y. 173, 638-640. STERLING, P. & WICKELGREN, B. (1969). Visual receptive fields in the superior colliculus of the cat. J. Neurophyeiol. 32, 1-23. STRYKER, M. & SHERK, H. (1975). Modification of cortical orientation-selectivity in the cat by restricted visual experience: a re-examination. Science, N.Y. 190, 904-906. TRETTER, F., CYNADER, M. & SINGER, W. (1975). Modification of direction selectivity of neurons in tht visual cortex of kittens. Brain Re8. 84, 143-149. VAN SLTnRERs, R. C. & BLAKEMoRE, C. (1973). Experimental creation of unusual neuronal properties in visual cortex of kitten. Nature, Lond. 246, 506-508. VITAL-DuiAND, F. & JEANNEROD, 'M. (1974a). Maturation of the optokinetic response: genetic and environmental factors. Brain Res. 71, 244-257. VIALnDuiAND, F. & JEANNEROD, M. (1974b). Role of visual experience in the development of optokinetic response in kittens. Expl Brain Re8. 20, 297-302. WIESEL, T. N. & HUBEL, D. H. (1963). Single responses in striate cortex of kittens deprived of vision in one eye. J. Neurophysiol. 26, 100-118. WIESEL, T. N. & Hu-BEL, D. H. (1965). Comparison of the effects of unilateral and bilateral eye closure in cortical unit responses in kittens. J. Neurophysiol. 28, 1029-1040. WYATT, H. J. & DAW, N. W. (1975). Directionally sensitive ganglion cells in the rabbit retina: specificity for stimulus direction, size and speed. J. Neurophysiol. 38, 613-626.