University of California Los Angeles California

Document Sample
University of California Los Angeles California Powered By Docstoc
					J. Physiol. (1978), 274, pp. 601-619                                                     601
With 10 text-ftgurem
Printed in Great Britain


    From the Beckman Laboratories of Behavioural Biology, California Institute of
     Technology, Pasadena, California 91125, and the Department of Physiology,
           University of California, Los Angeles, California 90024, U.S.A.
                                  (Received 17 January 1977)
   The reversibility of monocular deprivation was tested physiologically in twelve
kittens which had received varying amounts of normal visual experience prior to
unilateral eye closure. Recordings obtained from cells in cortical area 17 indicated
that the effects of monocular deprivation are reversible in cats as old as 8 weeks if
the animals have been allowed normal visual experience before eye closure.
    1. Four kittens had their right eyes closed at 3, 4, and 5 weeks of age. All animals
were reverse sutured at about 8 weeks of age, and after 2 weeks forced usage of the
deprived eye, each animal was tested physiologically for recovery. Single unit
recordings revealed a consistent relation between the extent to which the effects of
initial deprivation could reverse and the age of the animal at the time of this
deprivation. Recordings from a fifth, 8 week old kitten, which had been monocularly
deprived for 3 weeks, ensured that our shortest period of deprivation induces
a severe deficit.
   2. Two kittens which were reared in the dark for 4 and 5 weeks before unilateral
eye closure, failed to show substantial (greater than 20%) reversal when reverse
lid-sutured at 8 weeks of age. This failure suggests that kittens having their eyes
closed after normal experience show enhanced reversibility, at least in part, on
account of their early experience and not necessarily because of shorter deprivation.
A litter-mate of one of these kittens was allowed 9 days of normal vision after dark
rearing and before lid suture. Recordings which were performed both before and
after 6 days of reverse suture demonstrated a near total (85 %) reversal in ocular
   3. Three animals, monocularly deprived after some normal experience, recovered
substantial numbers of binocular cells after both eyes had been allowed to remain
open. In one of these cats the effectiveness of monocular deprivation was confirmed
physiologically before eye opening. These findings contrast with those of other workers
who have failed to find many binocular cells in kittens that had been allowed to
recover, with eye opening, from monocular deprivation which had lasted since
   4. One kitten was reared for 2 weeks (3-5 weeks of age) with alternating monocular
deprivation. Single unit recordings from this cat, which were made after four

                  Downloaded from J Physiol ( by guest on April 23, 2011
602             6. G. BLASDEL AND J. D. PETTIGREW
additional weeks of unilateral eye closure, revealed some cells which had resisted
monocular deprivation and which responded strongly to stimulation of the deprived
eye. Reverse suture in the same animal led to the recovery of many cells (36%),
dominated by the initially deprived eye, but did not result in a reversal in ocular
   The clinical condition of amblyopia ex anopsia has been recognized for some time
(Duke-Elder & Wybar, 1973). In 1963 and 1965 Hubel & Wiesel provided an experi-
mental analogue for this condition by subjecting kittens to several forms of visual
deprivation. They found that occlusion of one eye for six days markedly impairs that
eye's ability to influence cortical cells (Wiesel & Hubel, 1963; Hubel & Wiesel, 1970).
This impairment occurs if the eye suffers deprivation at any time during a 'sensitive
period' which, for the cat, lies between the third week and the third month of post-
natal life.
   The ability to recover substantially from the effects of monocular deprivation
seems to vanish well before the susceptibility to eye occlusion. Hubel & Wiesel (1970)
opened the sutured eye of a 5-week-old kitten which had been deprived since the
time of normal eye opening and, after waiting 14 weeks for the animal to recover,
found that only 25 % of the visually responsive cortical cells were dominated by the
initially deprived eye. Binocular cells, which in the normal adult comprise about
80 % of the cells in the visual cortex (Hubel & Wiesel, 1965), were conspicuously
absent. In another experiment they waited seven weeks before opening the deprived
eye and found almost no recovery. Blakemore & Van Sluyters (1974) achieved com-
parable results despite the use of reverse suturing where the animal is forced to use
its inexperienced eye because the lids of the initially exposed eye are sutured at the
same time that the lids of the deprived eye are parted. They found that although
reverse suturing before the fifth week led to a complete reversal in ocular dominance,
when the procedure was delayed until the eighth week, only 23 % of the cells reversed
their ocular dominance.
   These results imply that after 5 weeks of age the effects of monocular deprivation
quickly become irreversible. Hubel, Wiesel & LeVay (1976) have provided a possible
explanation for this irreversibility by noting anatomical changes in the bands of
geniculo-cortical terminals which represent each eye within layer IVc of monkey
visual cortex. These bands correspond to ocular dominance 'columns' and their
actual widths change by as much as 50 % following monocular deprivation. Bands
corresponding to the deprived eye are narrower and the territories which they would
normally occupy are filled by expanded bands of terminals from the non-deprived
eye. Since the difference in width is greatest in monkeys which are monocularly
deprived at the earliest ages, one might expect that the boundaries between these
bands would be largely determined during the rapid phase of visual cortical synapto-
genesis. Although the time course has not yet been determined for monkeys, in
kittens this occurs mostly between the eighth and thirty-seventh days of post-natal
development (Cragg, 1975a, b). Kittens are sensitive to monocular deprivation long
after 37 days of age, however; and one wonders whether monocular deprivation
which commences after synaptogenesis has occurred may produce effects which are

            Downloaded from J Physiol ( by guest on April 23, 2011
             REVERSIBILITY OF MONOCULAR DEPRIVATION                                 603
fundamentally different, and perhaps less enduring, from those produced by
monocular deprivation which has lasted from birth.
   We have explored this idea in three ways. We first exploited the finding that
animals which have been monocularly deprived from birth lose the ability to recover
substantially after 7 weeks of age (Hubel & Wiesel, 1970; Blakemore & Van Sluyters,
1974; Movshon, 1976). We provided kittens with normal binocular experience for
various lengths of time and then closed one eye. We assayed the effect which this
had on recovery by reverse suturing the animals at 8 weeks of age and making test
recordings 1 to 2 weeks later. Our results show that a kitten's ability to recover
increases with the amount of normal development which has been allowed before
eye occlusion.
   These results led us to try to distinguish between reversals in ocular dominance due
to reverse suturing and simple recovery of those cortical connexions which may
have become 'silent' or ineffective (Kratz, Spear & Smith, 1976; Duffy, Snodgrass,
Burchfiel & Conway, 1976). Since reversal necessitates some recovery of inputs
which have been either absent or non-functioning, animals which reverse their
ocular dominance are obviously capable of both processes. Measuring shifts of ocular
dominance after reverse suture, however, confuses the recovery of inputs from the
initially deprived eye with the concurrent suppression of inputs from the newly
sutured eye. The ability to suppress these inputs not only declines with age, but
probably also changes as inputs from the initially deprived eye recover. In an effort
to isolate 'pure recovery', we therefore adopted Hubel & Wiesel's technique of
simply opening the deprived eye. The results from this second group of experiments
suggest that, given previous binocular exposure, recovery also proceeds very well
when the initially exposed eye is allowed to remain open. In these cases, however,
there is no reversal in ocular dominance. Restored connexions from the deprived
eye become manifest as an increase in the proportion of binocular neurones.
   Finally, since we consistently found that normal binocular experience prior to lid
suture facilitated the recovery of inputs from the initially deprived eye, we tested
the way in which prior vision affects recovery if it occurs non-simultaneously for the
two eyes. We did this by alternately occluding the eyes of one kitten between 3 and
5 weeks of age and then suturing the lids of one eye for 4 additional weeks. Single
unit recordings which were performed after 3 weeks of reverse suture indicated that
this form of prior experience had also facilitated recovery.

   We recorded from single units in cortical area 17 of twelve kittens (taken from five separate
litters) each of which experienced monocular closure according to the protocol of Table 1.
Although the effects of monocular deprivation generalize to all kittens at early ages, it is possible
that the time course and extent of sensitivity to monocular deprivation are different for animals
of different ancestry. Such differences could account for variations in sensitivity to monocular
deprivation which have been observed for kittens from different litters (Hubel & Wiesel, 1970).
They would also contribute to the problem of comparing results from different laboratories where
recording techniques and rearing procedures may differ as well. In order to minimize this
variation, we took all but one of our litters from a partially inbred line of tabby cats. Data from
the litter (kittens with names containing the letter T) which originated outside our colony are
included because the conclusions which relate to kittens from this litter can be derived from

                  Downloaded from J Physiol ( by guest on April 23, 2011
604                     G. 0. BLASDEL AND J. D. PETTIGREW
comparisons between litter-mates. All kittens which experienced an illuminated environment
were exposed to light for 18 h per day.
   Initially it was always the right eye which was closed, the lids being sutured together under
halothane: N20 anaesthesia according to the technique of Wiesel & Hubel (1963). Following
surgery animals were returned to the colony and checked daily for the occurrence of 'windows'
(i.e. parting of the lid margins) in the sutured eyelids. In one kitten (T5); a window developed
following 1 week of reverse suture. Rather than repair the window, we altered the experimental
design for this kitten by separating the lids of the reverse sutured eye and allowing it 1 additional
week of unrestricted binocular vision. Results from animals which developed windows at any
time during either initial deprivation or reverse suture, are not included.

                    TABLE 1. Schedule of monocular deprivation and recording
                                      Right Right-           Left        1st     2nd Number
                  Rearing before       eye     eye           eye       record- record-   of
      Kitten           suture         closed opened         closed       ing     ing   units
        T1           light              21      53            53          66              39
        T2           light              28      55            55          68              29
        K2           light              28      54            54          67              36
        K3           light              35      56            56          69              57
        Ri           light              35      56                        56              36
        Ft           dark-28            28      53            54          53      60      61
        F2           dark-28            37      55            56          55      62      80
        L2           dark-37            37      63            63          70              34
        R4           light              35      42                        70              36
        T3           light              35      56          56-63         70              37
        Bi           light              35      45                        45      59      63
        T5         AMD* 21-35           35      64            65          64      84     71
        Total                                                                           579
   * Alternating monocular deprivation.
  This Table gives the ages (in days) of kittens at the time of each manipulation. The sequence
of kittens appearing in this table follows the order of their appearance in the Figures. This was
done to facilitate comparisons with the Figures. Kitten names are alpha-numerical and names
beginning with the same letter indicate animals from the same litter.
   We prepared for single unit recordings by using a variation of standard techniques (Barlow,
Blakemore & Pettigrew, 1967). After anaesthesia had been induced with 2% halothane in
a 2: 1 mixture of N20: 02, each animal was quickly intubated with a plastic tube (which had been
coated with xylocaine jelly) and cannulated with a 25 gauge butterfly needle in the radial vein.
We drilled a small hole over the left postlateral gyrus (stereotactic co-ordinates Pl-P3) and
reflected a small dural flap. Penetrations were always in the left hemisphere contralateral to the
initially deprived eye. An electrode (Levick, 1972) directed 200 anteriorly and 50 medially, was
then introduced through the crown of the postlateral gyrus and advanced 1-1-5 mm into the
medial aspect of the gyrus. Wound margins and pressure points were infiltrated with 2%
xylocaine. The animal was paralysed with Flaxedil (10 mg/kg. hr) contained in a 10% dextrose
solution and anaesthetized by hyperventilating with N20 (60-70 %) in an 02: C02 mixture. We
adjusted this mixture to keep end-tidal C02 between 3 and 3-5 %.
   The ocular alignment was estimated for each cat either by plotting the projections of the two
area centrale8 directly, or by plotting the projections of the two optic disks and using Bishop's
calculation (Nikara, Bishop & Pettigrew, 1968) to estimate the positions of the visual axes. The
most precise estimate of the alignment between the two eyes, however, was obtained from the
positions of receptive fields. Refraction was corrected during the plotting of the area centrale8
(Fernald & Chase, 1971) and optic disks by individually projecting the two retinae onto the
tangent screen and adjusting the convexity of the contact lenses until the projected images of
blood vessels became clear.
   Recording sessions were limited to 18 h. Dexamethasone (2 mg I.V.), atropine (2 mg i.V.) and

                Downloaded from J Physiol ( by guest on April 23, 2011
             REVERSIBILITY OF MONOCULAR DEPRIVATION                                           605
penicillin-streptomycin (I.M.) were injected at the beginning of each session. Since systemically
circulating atropine causes mild and symmetrical pupil dilation, artificial pupils were not neces-
sary. Small amounts of neosynephrine (1 %) were sometimes used to retract the nictitating
   After completion of a recording session, the animal was either allowed to recover or deeply
anaesthetized with Nembutal and perfused through the heart with saline followed by formol
saline. Brains from perfused animals were sectioned coronally at 40 #sm thickness on a freezing
microtome and stained with cresyl violet to allow for histological confirmation of the recording
   One half to one hour after positioning the electrode we slowly advanced it in a careful search
for units. When a unit was encountered, and its receptive field (minimum response field [Barlow
et al. 1967]) located we quickly determined three parameters - preferred orientation, orientation
specificity, and cell type (i.e. simple, complex, or hypercomplex) - for the dominant eye. We
then used this information, together with either the position of the non-dominant receptive
field or a knowledge of the ocular alignment and most probable position of this receptive field,
to determine the ocular dominance group of the cell (Hubel & Wiesel, 1962). After noting this
information we advanced the electrode 50 ,um before resuming the search for additional units.
We did this in order to distribute our sample of units over as large and regular a tangential
surface as possible and thereby to maximize the number of ocular dominance columns rep-
resented in a given sample. These columns are much less regular in the cat than they are in the
monkey (Schatz, Lindstrom & Wiesel, 1977) and the chances of sampling extensively from one
column during an 8 mm penetration are therefore minimal. When two units were resolved
simultaneously, we either recorded the more responsive one or, if they belonged to different
ocular dominance groups, we recorded both cells. We also noted the positions of units which we
could not stimulate; such unresponsive cells were rare, however, and did not appear to cluster.
   We tried to make our sampling procedure as standardized as possible. Our electrodes were
always of the same dimensions (12-15 ,um of exposed tungsten); tracks always began on the
medial aspect of the postlateral gyrus, 1-1 5 mm below the surface, and were angled in such away
(50 medially) as to distribute sampling across the deeper layers (6-3) of cortex. Receptive fields
were usually located in the visual field just below the projection of the area central and pro-
gressed slowly outward and downward with electrode advancement. Penetrations ended when
the receptive fields reached an eccentricity of 25°. Since the progression of receptive field
eccentricities increases rapidly near the periphery, most units were recorded within 15° of the
area centrali.

Recovery following reverse suture
   We found that the same period of reverse lid suture produced different amounts of
reversal in four kittens whose periods of initial monocular deprivation began at
different ages. This is shown in Figs. 1 and 2. Two litter-mates had their right eyes
closed at 3 and 4 weeks of age, respectively, and the kittens from a second pair at
4 and 5 weeks of age. All animals were reverse-sutured at approximately 8 weeks of
age. The results of single unit recordings, conducted after 2 additional weeks with
forced usage of the initially deprived eye, show a consistent relation between the
ages of the animals at the time of initial deprivation and the extent to which the
effects of this deprivation can reverse. While we found only one cell (out ofthirty-nine)
that was dominated by the initially closed eye of the kitten first deprived at 3 weeks,
55 % of the units sampled in its litter-mate, first sutured at 4 weeks ofage, had reversed
ocular dominance (Fig. 1). Similarly, one kitten from the second pair, first sutured
at 4 weeks of age, showed a 33 % reversal in ocular dominance and this was less than
the 65 % reversal shown by its litter-mate whose deprivation began at 5 weeks
(Fig. 2).
  This enhanced ability to restore inputs from the deprived eye is only interesting

                 Downloaded from J Physiol ( by guest on April 23, 2011
606                 G. G. BLASDEL AND J. D. PETTIGREW
if animals which become deprived at later ages show substantial deficits. That they
do so has been repeatedly demonstrated by previous experiments in which short
periods of deprivation, much shorter than we have used, reliably produce populations
of visual cortical cells which are strongly biased towards one eye (Hubel & Wiesel,
1970; Olson & Freeman, 1975). As further confirmation, however, 'we recorded from
one 8-week-old kitten whose right eye had been closed for 3 weeks. None of the cells
which we recorded were dominated by the deprived eye and only a few could even
be influenced by it (Fig. 3).
            40                 A                            20-                         B
                    T1                      35                        T2
                    n=39                                              n=29
                    _3%-                    -_-55%-


                          123456                                          12 34                     6

             0                                               0
                      1   2 3 4      5 6 7                                1   2 3 4          5      6 7

        Contra eye          32    13                    Contra eye                      27          13
           Ipsi eye -- B E G                               Ipsi eye
                   0 2 4 6 8                                          0       2     4 6         8        10
                      Age (weeks)                                                 Age (weeks)
      Fig. 1. Ocular dominance histograms for two litter-mates which had their right eyes
      closed at 3 (A) and 4 (B) weeks of age. Histograms are compiled from units which were
      recorded following 2 weeks of reverse suture. In contrast to the kitten which first had
      its right (contralateral) eye closed at an age of 3 weeks, its litter-mate, which underwent
      initial closure at 4 weeks, showed substantial reversal. Interestingly, 3-4 weeks is also
      the age when the optic media are beginning to clear (Thorn et al. 1976). Recordings
      were always made from the left hemisphere and units responding to the right eye are
      consequently contralateral. The convention for ocular dominance groups is the same as
      that introduced by Hubel & Wiesel (1962). Cells which respond exclusively to the contra-
      lateral (in this case, right) and ipsilateral (left) eyes fall into groups 1 and 7 respectively.
      Group 4 cells respond equally to both eyes and groups 2, 3, and 5, 6, contain cells
      which are binocular but which are dominated to varying degrees by the contralateral and
      ipsilateral eyes. Scales below each histogram indicate the visual history for each animal
      prior to physiological recording.

   In an effort to satisfy ourselves that these findings did not result from individual
variations, we monocularly deprived two additional litter-mates and tested them
physiologically before, as well as after, the reverse suture operation. By rearing them
in a totally darkened room until they were 4 weeks old, we more precisely controlled
the onset of visual experience with sharp contours. This would normally occur
between the third and the fourth week of life when the optic media become clear
(Thorn, Gollender & Erickson, 1976). We then closed the right eye of one kitten and

                 Downloaded from J Physiol ( by guest on April 23, 2011
             REVERSIBILITY OF MONOCULAR DEPRIVATION                                   607
brought both animals into the light for 9 days before closing the right eye of the other.
When the kittens were 8 weeks old, we recorded from single units in both animals
and verified that they possessed comparable monocular deficits (Fig. 4). Each kitten
was then reverse sutured and allowed to use its initially deprived eye for 6 additional
days before another recording session. The results appear in Fig. 5. The kitten which
had been allowed prior normal vision showed much more reversal (85 %) than its
litter-mate (18 %). In addition, the electrode track reconstructions from this animal
(Fig. 6) reveal that reversal was not confined to small clusters of cells but occurred
more or less evenly across the extent of sampled cortex.

                K2       A                                     K3            B
                n =43                                          n =57
        40-         33%-                                40      3665%-


  Z                                                                          8
          0'-   12 3 4 5 6 7                                      1 2 3 4 5 6 7
            Right eye             * Left eye                 Right eye            o Left eye

   Contra eye             26     13                 Contra eye               21 13
      Ipsi eye        -         _                      Ipsi eye          I _I
               0 2 4 6 8 10                                     0 2 4 6 8 10
                    Age (weeks)                                       Age (weeks)
       Fig. 2. Ocular dominance distributions for two litter-mates which had their right
       (contralateral) eyes initially closed at 4 (A) and 5 (B) weeks of age. Recordings were
       made at about 10 weeks of age, after 2 weeks of reverse suture. Again, the kitten which
       was initially deprived at a later age (B) shows a greater proportion of units dominated
       by the initially deprived (contra) eye than its litter-mate (A). Both kittens showed
       greater reversibility than Blakemore & Van Sluyters (1974) and Movshon (1976) found
       for 8 week old kittens which had been monocularly deprived from birth.

   One possible explanation for the greater reversibility of ocular dominance in
kittens which have been given prior experience is that monocular deprivation might
be less damaging if it begins later in the critical period. We examined this possibility
by rearing one kitten in darkness for the first 37 days of life. Its experience was
similar to that of kitten F1 (Figs. 4A, 5A) except that its 26 days of monocular
deprivation began when it was 9 days older. If the postponed onset of initial
deprivation were particularly relevant to reversibility, one would expect this kitten
to show more reversal than F1. In fact, it showed the same amount of reversal (Fig. 7)
In the absence of visual input, 9 days of postponement have not enhanced subsequent

                    Downloaded from J Physiol ( by guest on April 23, 2011
608                     G. G. BLASDEL AND J. D. PETTIGREW
Recovery after eye opening
  As mentioned previously, it is desirable to distinguish between reversals in ocular
dominance and the simple recovery of inputs from the deprived eye. We tried to do
this by abandoning the reverse suture procedure and adopting Hubel & Wiesel's
technique of opening the deprived eye. We began with a 6-week-old kitten that had
been monocularly deprived for 1 week. Occlusion for 6 days during this stage of
development dramatically shifts the ocular dominance of most visual cortical
                                     n=36                                                       33



                         0   _

                                          1       2               3   4            5       6    7

                                         Right eye                                     0   Left eye

                       Contra eye                                         21
                          Ipsi eye            =           z
                                     I                ,       ,       ,        ,
                                     0       42     6        8
                                        Age (weeks)
      Fig. 3. Ocular dominance histogram for an 8-week-old kitten which had its right eye
      closed for 3 weeks. This histogram shows the ocular dominance shift, in favour of the
      open eye, which one would expect from the findings of Hubel & Wiesel (1970). It demon-
      strates the severity of the impairment from which the kittens in Figs. 1 and 2 have

neurones in favour of the non-deprived eye, and even a 3-day occlusion induces
a rather severe depletion of binocular cells (Hubel & Wiesel, 1970). However, after
allowing 4 weeks for recovery, we found a large number of powerful inputs from the
deprived eye. Except for a slightly higher than normal representation of (group 7)
cells which responded exclusively to the non-deprived, ipsilateral eye, the ocular
dominance histogram for this animal appeared surprisingly normal. This is shown in
Fig. 8A. Many (67 %) of the sampled cells were binocular.
   Again, in order to verify that we were dealing with the recovery of animals which
had initially possessed substantial deficits, we conducted test recordings in the visual
cortex of another kitten prior to eye opening. This animal's right eye had been closed

              Downloaded from J Physiol ( by guest on April 23, 2011
                 REVERSIBILITY OF MONOCULAR DEPRIVATION                                                                              609

           30-                                                                                  B                    30
                                    A                              30-
                  F1                                                         F2
                  n=27                                                       n=40
                                                          22                 -10%-



             I                  __

                      1     2   3       4    5       6    7                                                      6   7
                                                                                                9       18
 Contra eye                             26                     Contra eye
    Ipsi eye                                                      Ipsi eye                  _
                 0 1234 56 78                                                012        34(5678
                    Age (weeks)                                                 Age (weeks)

Fig. 4. Results of test penetrations in the visual cortices of two litter-mates following
initial period of monocular deprivation. Both cats were dark reared for the first 4 weeks
of life. Ft (A) was then lid-sutured and brought into the light. F2 (B) received 9 days of
normal visual experience before lid suture. These test penetrations were conducted in
order to insure that each cat had a strong monocular deficit prior to reverse suture. The
results of recordings performed 1 week after reverse suture appear in Fig. 5.

           30-                  A                                        30-    -85%-                        B
                 F1                                  25
                 -18 %-


 z                6
                                                                               _ 4 4                             3        3

                  1        2 34             567                                     1       2       3 4 5 6               7
                   26 6                                           Contra eye
Contra eyer
   Ipsi eyeIIr"                                                      Ipsi eye                            IT
         0 2 4 6 8                                                              0       2           4    6       8
            Age (weeks)                                                                  Age (weeks)
Fig. 5. Ocular dominance histograms for two litter-mates. Unlike Ft (A), F2 (B), which
received 9 days of binocular experience prior to lid suture, was capable of an 85 %
reversal in ocular dominance. This reversal appeared in 6 days. Ft's failure to reverse
also contrasts with the reversal seen for the two kittensinFigs. 1 Band 2 A. Like F1, these
kittens each received 26 and 27 days of monocular deprivation which began at 4 weeks
of age.
20                                                                                                                        PHIY 274

                          Downloaded from J Physiol ( by guest on April 23, 2011
610                 G. G. BLASDEL AND J. D. PETTIGREW
between 35 and 45 days of age. As shown in Fig. 9A, very few (4 %) of the units
sampled after this treatment were dominated by the deprived eye and only 26%
could be influenced by it. Following 2 weeks of binocular vision, however the initially
deprived eye was able to drive 70 % of the recorded cells and dominated 33% (see
Fig. 9B). Half the units were binocular, and many of these received balanced inputs
from the two eyes (group 4).
                           A                          B
                                   1 2 34 5 6 7            1 2 3 4 5 6 7
                               l            1Fl
                       1                              1-

                       2-                            2-

                       3-                            3-

                 0     4-                            4-

                       5-                            5-
                       6                              6-

                       7                              7-

                       8J                           8J
    Fig. 6. Track reconstructions of recordings performed before (A), and after (B) reverse
    suture of kitten F2 (see Figs. 4B and 5B). These reconstructions indicate the unit
    samples on which the histograms in Figs. 4B (A) and 5B (B) are based. They demon-
    strate that reversal was not restricted to small clusters of cells but occurred more or
    less evenly along the extent of sampled cortex.

   Recordings from a third kitten which had its right eye closed for 3 weeks beginning
at 5 weeks of age followed by 1 week of reverse suture and 1 week of normal binocular
vision gave a similar result. A high proportion (68 %) of the recorded cells received
binocular inputs (Fig. 8B). Except for the increased presence of (group 7) cells which
responded monocularly to the non-deprived eye, these recordings revealed little to
indicate abnormal visual experience during rearing. The ocular dominance dis-
tribution did not differ greatly from that of the kitten whose eye had been closed
for only 1 week (Fig. 8A).

            Downloaded from J Physiol ( by guest on April 23, 2011
             REVERSIBILITY OF MONOCULAR DEPRIVATION                                                611

The effect of prior monocular experience on recovery
   We wanted to establish whether it was necessary for recovery that the early vision
be simultaneous for each eye. We did this by first raising one kitten with alternating
monocular deprivation. An opaque occluder was alternately placed over each of
its eyes on successive days between 21 and 35 days of age. Such treatment, if it occurs
                                      n=34                                            24

                                                  18% -



                           0   -

                                          1       2    3     4       5       6             7

                                                                    26           7
                        Contra eye
                           Ipsi eye
                                      0       2        4      6          8       10
                                                      Age (weeks)
    Fig. 7. Ocular dominance distribution for a kitten which suffered 26 days of monocular
    deprivation after 37 days of dark rearing. Instead of receiving 9 days of binocular
    experience before lid suture, like the kitten in Fig. 5B, this kitten spent an extra 9 days
    in the dark. Except for the postponed onset of monocular deprivation, its experience
    was similar to that of kitten FI (Fig. 5A). It is clear from this histogram, however, that,
    in the absence of visual experience, this postponement has not aided the recovery
    process. The amount of reversal (proportion of units dominated by the initially deprived,
    contralateral eye) indicated for this kitten is the same as that shown in Fig. 5A for Fl.
    Its failure to reverse substantially suggests that the dramatic reversal seen for the
    kitten in Fig. 5B is due more to its 9 days of binocular visual experience than it is to the
    delayed onset of unilateral deprivation.

during the most sensitive part of the critical period, quickly leads to a near-total
loss in binocular cortical cells (Hubel & Wiesel, 1965; G. G. Blasdel & J. D. Pettigrew,
in preparation). Whereas 80 % of the units in a normal cat are binocular, only about
10 % of the cells in an alternately occluded cat retain effective binocular inputs. We
closed the right eye at the age of 35 days and made recordings 4 weeks later to test the

                   Downloaded from J Physiol ( by guest on April 23, 2011
612                 G. G. BLASDEL AND J. D. PETTIGREW
effectiveness of eye closure (Fig. 1OA). These recordings revealed an extensive shift.
The shift was not as complete, however, as that seen in Fig. 3 for a kitten which was
initially deprived at the same age and for a shorter duration. If active afferent
                                             A                                           B

                            R4                                              T3
                            n=3675 %                                        n=37      76
                15 -                       67%                                           68%
                                           1 )

                                                                                        10      Q


                    0   -
                             1 2 3 4 5             6   7                       1   2 3 4 5 6 7
                0                                                0-

                1                                                1

                2                                                2

               E 3-                                              3-

         ~0 b

                6                                                6-

                7                                                7-

                8                                                8-1

         Contra eye                                           Contra eye
            Ipsi eye., ,                                         Ipsi eye    .  . ,,      I

                    0 1 2            3   4 5 6 7 8 910                      0 1 2 3 4 5 6 7 8 910
                                 Age (weeks)                                     Age (weeks)
                                         Fig. 8. For legend   see    facing   page.

           Downloaded from J Physiol ( by guest on April 23, 2011
            REVERSIBILITY OF MONOCULAR DEPRIVATION                             613
inputs suppress inactive ones as Wiesel & Hubel (1965) have suggested, then this
incomplete shift may reflect the highly monocular state which some neurones have
achieved following 2 weeks of alternating monocular deprivation. Unaffected cells
may not be vulnerable to deprivation because they totally lack inputs from the
exposed eye.
  Following the first recording session, the kitten was allowed to recover with its
          20                 A                            20-                   B
                 n=23                  17n40
                          26%                                               70%
                           26%                                               48%
                        a)                                        ~~~~~~~~~~~~~9

                                 3                                          3

            0 _J0
                    1 2 3 4 5 6 7                                  1 2 3 4 5 6 7
                            10                                           10 14
      Contra eye        e                             Contra eye
         Ipsi eye,- , , , ,     ,                        Ipsi eye . , ,
                  0 2 4 6                                        0 2 4 6 8
                   Age (weeks)                                         Age (weeks)
    Fig. 9. The ocular dominance histograms in this Figure contain the results of pene-
    trations conducted in the same animal before (A) and after (B) eye opening. A, follow-
    ing 10 days of monocular deprivation, this histogram illustrates that most of the recorded
    cells were dominated by the non-deprived eye. B, allowing 2 weeks for recovery, sub.
    sequent penetrations revealed that the deprived eye had regained influence over 70 %
    of the recorded cells. Many of these cells were binocular. The somewhat unusual appear-
    ance of this histogram derives mostly from combining the results of two penetrations.
    The first of these revealed cells with receptive fields in the area centralis and these cells
    tended to be monocular, falling into ocular dominance groups 1 and 7. This is not an
    altogether unexpected result (see Albus, 1975). In the second penetration, which was
    located lower in the visual field, cells influenced by the deprived eye generally responded
    equally to either eye and hence fell into ocular dominance group 4.

    Fig. 8. Ocular dominance distributions for kittens which were allowed to use both eyes
    simultaneously after the initial period of eye closure. The kitten in A had its right eye
    closed for 7 days, beginning at 35 days of age. Hubel & Wiesel (1970) have previously
    shown that 6 days of unilateral eye closure at this time results in a drastic reduction in
    the number of binocular neurones and a nearly complete ocular dominance shift in
    favour of the non-occluded eye. After opening the occluded eye and allowing it to
    remain open for 4 weeks, we found that many (67 %) cells had recovered binocular
    inputs. This cat was not obviously strabismic, as judged by the approximately normal
    alignment of the visual axes in the paralysed state. The kitten in B had its right eye
    closed for 3 weeks beginning at 5 weeks of age. Following this period, it was reverse
    sutured for one week and thereby forced to use its initially deprived eye. Following this,
    both eyes were allowed to remain open for one additional week before recording. Results
    from this recording session indicate a return of binocularity which is comparable to
    that (seen following one week of closure) in A.

                  Downloaded from J Physiol ( by guest on April 23, 2011
614                 G. G. BLASDEL AND J. D. PETTICREW
initially deprived eye open. The initially exposed eye was then sutured shut. Allowing
19 days for reversal, we again made single unit recordings and found that, while
many units had recovered their responsiveness to the deprived eye, there was no

                                         A                                      B

              30 -                                 29
                                                                          _      55%

                             1   2 3    4    5    6 7                   1   2   3 4 5       6   7
              1   -                                        1-

              2   -
                                                   -       2-

              3   -


       I-c    5-


              7-                                           6-

              8-                                           7-

       Contra         eye                    29         Contra   eye                   29           19
          Ipsi        eye        aI                       Ipsi   eye            '               I

                            0 1 2 3 4 5 6 7 8 9                        0 1 2 3 4 5 6 7 8 91011
                                 Age (weeks)                                    Age (weeks)

                                       Fig. 10. For legend   see   facing   page.

              Downloaded from J Physiol ( by guest on April 23, 2011
            REVERSIBILITY OF MONOCULAR DEPRIVATION                                  615
apparent suppression of inputs from the initially exposed eye (Fig. lOB). Many units
responded to each eye and the cortex appeared to be returning to a state in which it
was evenly divided between representations for the two eyes. This limited amount
of reversal seems to be a simple case of recovery, and suggests that binocular
experience before lid suture need not be simultaneous in order to facilitate subsequent
recovery. Failure to reverse entirely may result from two factors. The difficulty of
suppressing cells that do not receive inputs from the exposed eye has been mentioned.
In addition, reverse suturing in this animal did not occur before the ninth week and,
even though recovery might still be possible, the ability to suppress deprived inputs
is declining rapidly at this time.
  Cragg's data (1975), show that synapse formation in the developing primary visual
cortex of the kitten nears completion around the fifth week of life. It seems likely
that deprivation which occurs before this time may produce fundamental and enduring
rearrangements of cortical architecture more readily than later deprivation. Since a
strong deficit can still be produced after 5 weeks, when neuropil development is rela-
tively complete, and since eye closure at this time does not cause fibre degeneration
(Cragg, 1972), shifts in ocular dominance occurring later may result from some mechan-
ism other than large scale structural changes. This suggestion is supported by the
present findings to the extent that substantial recovery and reversal are possible at
quite a late age if a cat has had normal experience before monocular deprivation.
Such recovery is not possible, however, for animals that have been monocularly de-
prived since birth or even for animals which have been monocularly deprived for
shorter periods but which have not received previous experience (i.e. which have been
dark-reared). In addition, this experience appears to help only if it occurs after about
3 weeks of age, when the optic media are beginning to clear (Thorn et al. 1976).
   Delaying the onset of monocular deprivation does not itself improve the chance
of getting reversal of ocular dominance since the kitten whose initial deprivation was
delayed until 37 days after birth, but which remained in the dark during this time,
failed to show much reversal (Fig. 6B). The duration of lid suture may also be
important, but again, by itself, this cannot explain the results either. The four

    Fig. 10. We raised one kitten with alternating monocular deprivation before lid suture
    in order to test whether the experience before deprivation (experience that enables
    subsequent recovery) needs to occur simultaneously for the two eyes. A, following
    4 weeks of monocular deprivation, which began at 5 weeks of age, single unit recordings
    indicated that the resulting ocular dominance shift, though extensive, was not complete.
    B, following 2 weeks of reverse suture, the percentage of sampled cells dominated by the
    initially sutured eye grew from 11% to 36 %. There was no indication, however, that
    inputs from the initially dominant eye became suppressed during reverse suture. Since
    most of this cat's visual cortical cells were monocular (on account of the alternating
    monocular deprivation) before lid suture, the difficulty of suppressing cells which
    respond exclusively to the sutured eye, and the recovery of these cells, which went
    unaccompanied by the disappearance of cells dominated by the other eye, both suggest
    that monocular deprivation has less effect on cells which are already strongly domi-
    nated by one eye. This suggestion is in accordance with the predictions of Hubel &
    Wiesel (1970).

                Downloaded from J Physiol ( by guest on April 23, 2011
 616               6. 0. BLASDEL AND J. D. PETTIGREW
kittens in Figs. 1B, 2A, 5A, 7 were all subject to four weeks of monocular
deprivation, yet only those animals with early binocular experience (Figs. 1 B, 2A)
showed significant subsequent reversal. This interpretation is made somewhat
awkward by the fact that these kittens were not all allowed the same amount of
time in which to recover. As Movshon (1976) has pointed out, however, most of the
recovery that occurs does so quickly and is apparent within 6 days of reverse suture,
which was the minimum time allotted to any of our animals. This was certainly
enough time for one of our kittens (Figs. 4B, 5B) to show a near total reversal.
   There are a number of reasons why the cortical cells in our cats may have succeeded
in regaining binocular connexions while those of Hubel & Wiesel's (1970) cats failed.
One possible explanation is that the normal binocular experience which our kittens
received before lid suture somehow strengthened binocular inputs and thereby
decreased the vulnerability of these inputs to destruction. Our kittens also had the
opportunity to develop proper ocular alignment during the first 5 weeks of life; and
correct alignment, which was apparent at the time of recording, may have facilitated
the return of binocular connexions. Kittens which do not have this opportunity
during the first weeks of vision, because of either dark-rearing or lid suture, develop
a strabismus (Pettigrew, 1974; Kalil, 1975), and this, by itself, can reduce binocu-
larity (Hubel & Wiesel, 1965). In addition, strabismus may hamper the recovery of
inputs from the deprived eye.
   It is particularly intriguing that binocular connexions return with such strength
under the appropriate conditions. Their return reveals binocularity to be less fragile
than has been thought (Hubel & Wiesel, 1970) and suggests that the effects of
monocular deprivation occurring later in the critical period may be at least initially
transient. Even 1 week of monocular deprivation produces permanent effects,
however, and these are reflected in the somewhat higher than normal proportion of
(group 7) units which respond exclusively to the initially non-deprived eye (see for
example Fig. 8A). Although it is conceivable that longer recovery periods would
reduce this proportion, such recovery must be slow. The predominance of group 7
cells was apparent even though the cat had 4 weeks in which to recover. The persist-
ence of this effect suggests that degenerative changes have occurred. Such changes
may have participated in the recovery of all our animals although the extent of this
participation is not yet clear. It is possible that longer durations of monocular
deprivation (greater than 3 weeks) might lead to less reversible effects, even though
the cortex had developed normally through the early period of rapid synaptogenesis.
   The lack of a complete ocular dominance shift in the kitten whose eyes had been
alternately occluded before 4 weeks of unilateral lid suture (Fig. IOA) was somewhat
surprising. It is possible that 2 weeks of alternating monocular deprivation produced
visual cortical neurones which were so strongly monocular that they resisted depri-
vation. If so, it supports the suggestion (Wiesel & Hubel, 1965; Hubel & Wiesel,
1970; Sherman, Guillery, Kaas & Sanderson, 1974) that neurones shift ocular domi-
nance in response to competition between convergent binocular inputs. If inputs
from the non-deprived eye are absent for a particular group of cells, then inputs from
the deprived eye may not be vulnerable. This result would then be analogous to that
of Sherman et al. (1974) who removed the influence of the exposed eye from a small
group of cortical cells by making retinal lesions. It might also explain the few

            Downloaded from J Physiol ( by guest on April 23, 2011
             REVERSIBILITY OF MONOCULAR DEPRIVATION                                  617
unaffected cells in the first recordings from our binocularly experienced dark-reared
cat (Fig. 4B). Since dark-rearing induces strabismus (Pettigrew, 1974; Kalil, 1975)
it may have led to an abnormally high percentage of monocular cortical cells (Hubel
& Wiesel, 1965) during this kitten's first binocular exposure at 4 weeks of age.
However, the findings of Stryker & Schatz (1976) should make one wary of placing
too much emphasis on small numbers of unaffected cells. Even in a cat which has
undergone long-term monocular deprivation, these authors find that there are some
cells (largely in layer IVc) which respond exclusively to the deprived eye.
   The idea that deprivation, which occurs later in the critical period, expresses itself
in a fundamentally different and less enduring manner has received additional recent
support. Olson (personal communication) has independently observed the return of
binocularity in cells following eye opening. Blakemore et al. (1976) report that the
orientation preferences of cells responding to a reverse-sutured eye appear more in
register with the orientation preferences of cells responding to the other eye if cats
receive binocular experience prior to lid suture. They hypothesize that the orientation
preference of a cell, once established, will remain even though the inputs from one
eye are made temporarily non-functional. Finally, Hubel, Wiesel & LeVay (1977),
who observed that ocular dominance 'columns' change width in response to monocular
deprivation, found that this reduction in width of 'columns' corresponding to the
deprived eye was greater (47 %) in a monkey monocularly deprived at an age of
2 weeks than it was (33 %) in a monkey deprived at 3 weeks. Deprivation after
6 weeks of age produced no change at all (Hubel et al. 1977), even though we may
presume that physiological changes occurred (Baker, Grigg & Van Noorden, 1974).
Interestingly, deprivation amblyopia which is induced after 6-8 weeks of age has
been found to be reversible in behavioural experiments (Von Noorden, 1972).
   In conclusion, the primary visual cortex may respond to monocular deprivation
in two phases. One phase could correspond to the period of rapid synaptogenesis when
terminals corresponding to the two eyes compete for cortical space and the boundaries
of ocular dominance 'columns' become defined. The second phase, then, would occur
as rapid synaptogenesis subsides. The ocular dominance of many neurones still
responds to monocular deprivation, but the boundaries between 'columns', having
once been defined, are not easily redefined. Effects that develop during the first
phase may derive from large scale anatomical abnormalities and, if so, are probably
irreversible. Effects that develop during the second phase, on the other hand, prob-
ably do not involve wholesale structural changes. These are apparently characterized
by greater reversibility.
   This work was supported by a grant from the Spencer Foundation and U.S. Public Health
Service grant MH 25852 to J.D.P. We thank Herb Adams for excellent technical assistance and
Sarah Kennedy for assistance during the surgery, histology, and some of the initial recording
sessions. In addition we thank Drs Alan Grinnell and Jim Hudspeth for helpful suggestions.

                 Downloaded from J Physiol ( by guest on April 23, 2011
618                   G. G. BLASDEL AND J. D. PETTIGREW

ALBUS, K. (1975). Predominance of monocularly driven cells in the projection area of the
  central visual field in cat's striate cortex. Brain Res. 89, 341-347.
BAKER, F. H., GRIGG, P. & VON NOORDEN, G. K. (1974). Effects of visual deprivation and
  strabismus on the response of neurons in the visual cortex of the monkey, including studies on
  the striate and prestriate cortex in the normal animal. Brain Res. 66, 185-208.
BARLOW, H. B., BLAKEMORE, C. & PETrIGREW, J. D. (1967). The neural mechanism of binocular
  depth discrimination. J. Physiol. 193, 327-342.
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. & MovsHoN, J. A. (1976). Synaptic competition in the
  kitten's visual cortex. The synapse. Cold Spring Harb. Symp. quant. Biol. 40, 601-609.
CRAGG, B. G. (1972). The development of synapses in cat visual cortex. Invest. Ophth. 11,
CRAGG, B. G. (1975a). The development of synapses in the visual system of the cat. J. comp.
  Neurol. 160, 147-166.
CRAGG, B. G. (1975b). The development of synapses in kitten visual cortex during visual
  deprivation. Expl Neurol. 46, 445-541.
DUFFY, F. H., SNODGRASS, S. R., BURCHFIEL, J. L. & CONWAY, J. L. (1976). Bicuculline reversal
  of deprivation amblyopia in the cat. Nature, Lond. 260, 256-257.
DuKE-ELDER, S. & WYBAR, K. (1973). Ocular motility and strabismus. In System of Ophthal-
  mology, vol. 6, ed. Duke-Elder, S., pp. 290-315. London: Henry Kimpton.
FERNALD, R. & CHASE, R. (1971). An improved method for plotting retinal landmarks and
  focusing the eyes. Vision Res. 11, 95-96.
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. (1965). Binocular interaction in striate cortex ofkittens reared
  with artificial squint. J. Neurophyoiol. 28, 1041-1059.
HUBEL, D. H. & WIESEL, T. N. (1970). The period of susceptibility to the physiological effects
  of unilateral eye closure in kittens. J. Physiol. 206, 419-436.
HUBEL, D. H., WIESEL, T. N. & LEVAY, S. (1976). Functional architecture of area 17 in normal
  and monocularly deprived macaque monkeys. Cold Spring Harb. Symp. quant. Biol. 40,
HUBEL, D. H., WIESEL, T. N., & LEVAY, S. (1977). Plasticity of ocular dominance columns in
  monkey striate cortex. Phil. Trans. R. Soc. B. 278, 377-409.
KALIL, R. E. (1975). Recovery from dark rearing: Behavioural and anatomical observations.
  Neurosci. Abstr. 1, 85.
KRATZ, E., SPEAR, P. D. & SMITH, D. C. (1976). Postcritical-period reversal of effects of monocular
  deprivation on striate cortex cells in the cat. J. Neurophysiol. 39, 501-511.
LEVICK, W. R. (1972). Another tungsten microelectrode. Med. Electron. & biol. Engng 10,510-515.
MovsHoN, J. A. (1976). Reversal of the physiological effects of monocular deprivation in the
  kitten's visual cortex. J. Physiol. 261, 125-174.
NIKARA, T., BISHOP, P. 0. & PETTIGREW, J. D. (1968). Analysis of retinal correspondence by
  studying receptive fields of binocular single units in cat striate cortex. Expl Brain Res. 6,
OLSON, C. R. & FREEMAN, R. D. (1975). Progressive changes in kitten striate cortex during
  monocular vision. J. Neurophysiol. 38, 26-32.
PETTIGREW, J. D. (1974). The effect of visual experience on the development of stimulus speci-
  ficity by kitten cortical neurones. J. Physiol. 237, 49-74.
SCHATZ, C. J., LINDSTROM, S. H. & WIESEL, T. N. (1977). The distribution of afferents represent-
  ing the right and left eyes in the cat's visual cortex. Brain Res. 131, 103-116.
SHERMAN, S. M., GUILLERY, R. W., KAAS, J. H. & SANDERSON, K. J. (1974). Behavioural,
  electrophysiological and morphological studies of binocular competition in the development
  of the geniculo-cortical pathways of cats. J. comp. Neurol. 158, 1-18.
STRYKER, M. P. & SCHATZ, C. J. (1976). Ocular dominance in layer IV of the normal and deprived
  cat's visual cortex. Neurosci. Abstr. 2, 1137.

             Downloaded from J Physiol ( by guest on April 23, 2011
             REVERSIBILITY OF MONOCULAR DEPRIVATION                                         619
THORN, F., GOLLENDER, M. & ERICxKsoN, P. (1976). The development of the kitten visual optics.
  Vi8ion Re8. 16, 1145-1149.
VON NOORDEN, G. K. (1972). Experimental amblyopia. Israel J. med. Sci. 8, 1496-1499.
WIESEL, T. N. & HUBEL, D. H. (1963). Single cell responses in striate cortex of kittens deprived
  of vision in one eye. J. Neurophypiol. 26, 1003-1007.
WIESEL, T. N. & HUBEL, D. H. (1965). Comparison of the effects of unilateral and bilateral eye
  closure on cortical unit responses in kittens. J. Neurophyeiol. 28, 1029-1040.

                Downloaded from J Physiol ( by guest on April 23, 2011