European Journal of Neuroscience, Vol. 11, pp.

1301–1312, 1999 © European Neuroscience Association

Auditory activation of cortical visual areas in cats after

early visual deprivation

Rami Yaka,1,2 Uri Yinon2 and Zvi Wollberg1

1Department of Zoology, George. S. Wise Faculty of Life Sciences, Sheba Medical Center, Sackler Faculty of Medicine,

Tel-Aviv University, Tel-Aviv 69978, Israel

2Goldschleger Eye Research Institute, Sheba Medical Center, Sackler Faculty of Medicine, Tel-Aviv University, Tel-Aviv 69978,

Israel

Keywords: cortex, cross-modal neuroplasticity, enucleation, suprasylvian sulcus, visual areas

Abstract
Auditory activation of the primary visual cortex (area 17) and two extrastriate visual cortical areas – the anterolateral lateral
suprasylvian area (ALLS) and anteromedial lateral suprasylvian area (AMLS), was investigated in visually impaired cats. Impairment
was accomplished shortly after birth by bilateral eyelid suturing (binocularly deprived cats, BD) or bilateral enucleation (binocularly
enucleated cats, BE). In BE cats, the optic nerve and chiasm were entirely degenerated. No cortical atrophy or cytoarchitectural
malformation was noticed in either BD or BE cats. In both normal and impaired cats we found auditory-responsive cells in the
ALLS and AMLS, areas that are considered strictly visual. The most remarkable finding was an increase in the relative number of
these auditory cells in the BD and BE cats, which was more prominent in the latter. Some auditory-responsive cells were also
found in area 17 of BE cats. On the basis of formal calculation, it is tempting to suggest that the increase in relative number of
auditory cells in these areas reflects the transformation of all the visual cells in the ALLS of BD and BE cats into auditory cells. In
BE cats, all bimodal cells and an appreciable percentage of non-responsive cells also had transformed to auditory cells. In the
AMLS of BD cats, it is primarily the bimodal cells that become auditory cells, whereas in BE cats all the visual and bimodal cells
as well as non-responsive cells undergo this transformation. This assumption, however, is one possible interpretation of our results
but not the only one. Other modes of neuronal plasticity that might yield similar results in the visually deprived cats can not be
ruled out.

Introduction
Early visual deprivation in mammals, especially in kittens, has been
widely used for the past three decades for studying intramodal
neuroplasticity (Hubel & Wiesel, 1970; Kasamatsu & Adey, 1974;
Yinon, 1980; Spear et al., 1983). Many of these studies focused on
structural and functional reorganization of cortical visual areas,
primarily the striate cortex, following visual deprivation. It has been
shown that cats reared as kittens in the dark (Yinon & Goshen, 1984),
or with their eyelids sutured (Spear et al., 1983), possess a relatively
smaller number of visually driven cells. In addition, the remnant
responding cells in these animals were less selective to various visual
stimuli. These findings raise questions as to the possible role of the
cells that lost their sensitivity to visual stimuli following early visual
deprivation, and the fate of visual areas in completely blind animals.
One very tempting possibility is that these areas become activated
by other sensory modalities.
Data concerning such possible cross-modal neuroplasticity are quite
sparse, and until the last decade mostly came from sporadic clinical
reports and psychophysical examinations of human subjects. Experi-
ments with young adults have shown that blind subjects possess
certain superior auditory capabilities (Kellogg, 1962; Woods et al.,
1985), including superior discrimination abilities of speech sounds,
probably reflecting neural reorganization at the upper levels of the
Correspondence: Z. Wollberg, as above. E-mail: wollberg@ccsg.tau.ac.il
Received 7 April 1998, revised 4 November 1998, accepted 10 November 1998
auditory system (Niemeyer & Starlinger, 1981). More recent studies
have shown that during the performance of auditory and/or somatosen-
sory tasks, increased metabolic activity and enhanced regional blood
flow in the occipital cortex of early blind humans occurs (De Volder
et al., 1997), but not in subjects who lost their sight later on in their
development (Veraart et al., 1990). These findings have recently been
further corroborated by recording electrical event-related responses,
MEG and PET studies (Uhl et al., 1991; Alho et al., 1993; Kujala
et al., 1995; Sadato et al., 1996), which suggest that cortical visual
areas in early blinded humans are involved in the processing of
auditory and/or somatosensory information.
The findings in visually impaired humans are consistent with data
from experimental animal models that constitute most of the available
physiological reports. For instance, it has been shown that in neonatally
enucleated mice, representation of the whisker pad in the primary
somatosensory cortex (the barrel field) expands into the visually
deprived occipital cortex (Toldi et al., 1994). Neuronal and behavi-
oural auditory compensation have been demonstrated in cats
binocularly deprived at birth by eyelid suturing (Rauschecker &
Korte, 1993, Rauschecker & Kniepert, 1994). In such cats, auditory-
responsive cells have been found in the superficial layers of the
superior colliculus (SC) (Rauschecker & Harris, 1983), which in intact
animals are exclusively visual. In addition, cells in the intermediate and
deep layers of the SC and the anterior ectosylvian cortex, which in
intact animals are multimodal or primarily visual, responded only to
auditory and somatosensory stimuli. Auditory spatial tuning of single
1302 R. Yaka et al.

FIG. 1. Recording areas. (A) Top view of the cat’s brain. Rectangles delineate the primary visual cortex (A17) and the suprasylvian region including the non-
primary visual fields and the auditory fields. The lines A2, P6 in A17, and A13, A9 in the suprasylvian region, designate the anterior and posterior borders of
the recording areas following the frontal coordinates of the Horsley–Clarke stereotaxic system. (B,D) Enlargements of the corresponding rectangles in (A). The
vertical lines indicate the plans of a coronal section shown in (C) and (E). (C,E) Drawings of coronal sections through the suprasylvian region and A17
illustrating the recording sites. ALLS, anterolateral lateral suprasylvian area; AMLS, anteromedial lateral suprasylvian area; AI, primary auditory cortex; AII,
secondary auditory cortex; AAF, anterior auditory field; SSS, suprasylvian sulcus; AES, anterior ectosylvian sulcus; PES, posterior ectosylvian sulcus; PLS,
post-lateral sulcus.

cells in the anterior ectosylvian cortex of visually deprived cats was
significantly sharper than that of normal cats (Korte & Rauschecker,
1993). A behavioural correlate, apparently manifesting these physio-
logical findings, was that of the enhanced auditory localization
capabilities exhibited by these deprived cats (Rauschecker & Kniepert,
1994). Electrophysiological and 2-deoxyglucose studies in Spalax
ehrenbergi, a congenitally blind subterranean rodent, revealed that
primary visual areas, e.g. the dorsal lateral geniculate nucleus (dLGN)
and the striate cortex, are activated by auditory stimuli (Bronchti
et al., 1989; Heil et al., 1991). The major source of auditory input to
these visual areas has been shown to be the inferior colliculus, which
in addition to all its typical auditory targets, also projects to the
dLGN (Doron & Wollberg, 1994). It has also been shown that
transient somatosensory projections which invade the dLGN of
normal mice are consolidated in congenitally anophthalmic adult
mice (Asanuma & Stanfield, 1990). It is thus evident that visual
impairment can induce cross-modal neuronal reorganization that is
also correlated with some behaviourally manifested superior hearing
capabilities.
Complete blindness from birth is a known pathological impairment
in humans. Sutured eyelids, however, the most common experimental
model for visual deprivation, still enables penetration of diffused
light. It is thus of interest to determine how blindness from birth
affects the reorganization of the visual system. The main goal of this
study was to look for possible auditory activation of the cortical
striate and extrastriate visual areas in early enucleated cats and in
cats that were visually deprived by the routine procedure of eyelid
suturing, and to compare the two. Intact cats were used as controls.
Three visual areas were selected for this purpose: the primary
visual cortex (A17), anterolateral lateral suprasylvian area (ALLS)
and anteromedial lateral suprasylvian area (AMLS). The ALLS and
AMLS, two extrastriate areas, are located in the lateral and medial
banks of the suprasylvian sulcus (SSS), respectively, and are typically
considered as strictly visual (Clare & Bishop, 1954; Hubel & Wiesel,
1969; Spear & Bauman, 1975; Palmer et al., 1978). These areas were
first described by Marshal et al. (1943), Clare & Bishop (1954), and
Hubel & Wiesel (1969), and electrophysiologically found to be
retinotopically organized (Spear & Bauman, 1975; Turlejski &
Michalski, 1975; Palmer et al., 1978). Their connections with the
cortical and thalamic areas have been quite extensively studied and
documented (see review by Scannell et al., 1995). In both the normal
and impaired cats, in addition to visual cells, we also found auditory
and bimodal cells. Our main finding, however, was a prominent
activation of the two extrastriate visual areas by auditory input, the
degree of which depended on the severity of deprivation. In addition,
the enucleated cats also demonstrated some auditory activity in
area A17.
Preliminary results have been previously published in abstract form
(Yaka et al., 1996).
Materials and methods
Animal preparations
A total of 28 cats were used in this study. Eight cats were visually
deprived by binocular eyelid suturing (binocularly deprived cats,
BDs); seven were binocularly enucleated by complete removal of the
eyes (binocularly enucleated cats, BEs); and 13 served as controls
(NORs). Both types of deprivation were performed 3–5 days after
birth under deep anaesthesia (ketamine hydrochloride, 25 mg/kg,
i.m.). Normally, kittens open their eyes about µ 10–11 days after
birth. To minimize postoperative pain, the area around the eyes was
treated before the operation with a long-lasting local anaesthetic
(lidocaine hydrochloride, 2%). Prophylactic antibiotic treatment con-
sisted of synthomycine ointment (5%) on the skin, gentamicin
(16 mg/kg/day, i.m.) and synthomycine (200 mg/kg/day, i.m). Detailed

© 1999 European Neuroscience Association, European Journal of Neuroscience, 11, 1301–1312


FIG. 2. Whole brain of a normal cat (NOR) (A; A1); of a cat with both eyelids sutured (BD) (B; B1); and of a binocularly enucleated cat (BE) (C; C1). (A–C)
Dorsal aspects; (A1–C1) ventral aspects. Note the complete degeneration of the optic nerve and chiasm in the enucleated cat (arrow in C1).
suturing procedures have been previously described (Yinon et al.,
1993). Upon awakening from the anaesthesia, the operated kittens
were returned to their mothers until complete weaning. Thereafter,
until the recording session (1.5–2 years later), each cat was held
individually in a separate cage.
All surgical and experimental procedures were performed in accord-
ance with the standards for care and use of laboratory animals and
has been approved by the TAU Animal Care and Use Committee
(approval L 95-001).
Surgery procedures prior to the recording session were performed
under deep anaesthesia achieved by i.m. administration of ketamine
hydrochloride (15 mg/kg), xylazine hydrochloride (2 mg/kg) and
atropine sulphate (0.01 mg/kg). The anaesthetized animals were then
tracheotomized and placed in a stereotaxic apparatus using hollow
ear bars. Subsequently, and during the entire recording session,
animals were kept under light anaesthesia (thiopental sodium, 3 mg/
h i.m). The eyelids of the BD cats were reopened at the onset of the
recording session, immediately following anaesthesia.
To avoid eye movements while mapping visual receptive fields,
the animals were also paralysed by a continuous infusion of gallamine
triethiodide (7.5 mg/kg/h via the cephalic vein) and artificially respir-
ated with a rate- and volume-calibrated animal respirator (Harvard
Apparatus model 661, MA, USA). The cats were also infused
continuously with atropine (0.03 mg/kg/day), potassium chloride
(4.0 mg/kg/h), isoproterenol hydrochloride (isuprel 6 µg/kg/h) and
dextrose–saline solution (2 ml/kg/h). Body temperature was main-
tained by means of a heating pad and a DC temperature control module
(Ealing, Kent, UK). The ECG was displayed on an oscilloscope and
audibly monitored. Urine was collected into a measuring bag and
expiratory CO2 levels were measured using a gas analyser (Beckman
LB2, IL, USA). All were constantly monitored and kept within the
physiological range. Saline solution was continuously applied to the
cornea to prevent drying.
Electrode penetration sites were accessed by craniotomy and the
removal of the dura matter. To approach the ALLS and AMLS, the
© 1999 European Neuroscience Association, European Journal of Neuroscience, 11, 1301–1312
Auditory activation of cortical visual areas 1303
bone was removed between Horsley–Clarke coordinates A9 and A13.
To approach the primary visual cortex (A17), an aperture was made
between P6 and A2 (Fig. 1).
Recording and stimulation procedures
While the borders of the cat primary visual cortex (A17) are well
documented (Otsuka & Hassler, 1962; Woolsey, 1971), neither the
border between the primary auditory cortex (A1) within the lateral
bank of the SSS and the dorsal part of the ALLS (Palmer et al.,
1978), nor the border between the suprasylvian gyrus and the AMLS
have been distinctly defined. Our own attempts to determine these
borders by cytoarchitectonic criteria, using Nissl stain, also yielded
ambiguous results. Thus, in normal and BD cats we defined these
borders electrophysiologically by the appearance of the first visually
driven cell. The mean depth (6 SD) that such cells occurred at in
the ALLS of normal cats was 4.5 6 0.7 mm from the cortical surface,
and the corresponding value in BD cats was 4.8 6 0.7 mm. These
two values do not differ significantly (unpaired t-test; P . 0.5) and
are in accordance with the border line of the primary auditory cortex,
as previously determined (Imig & Reale, 1980). By the same criterion,
the dorsal border line of the AMLS in normal cats was at a depth of
1 6 0.3 mm from the cortical surface and 1.6 6 0.3 mm in BD cats,
with no significant difference between the two (unpaired t-test;
P . 0.1). Because BE cats do not possess visually driven cells, to
determine the border lines in these animals we relied on the mean
values found in normal and BD cats, assuming that these borders
also fit the enucleated animals.
Single unit activity was extracellularly recorded with epoxylite-
coated tungsten microelectrodes (impedance: 4–5 MΩ at 1 kHz)
advanced perpendicularly to the cortical surface by means of an
hydraulic microdrive (David Kopf, CA, USA). Action potentials were
amplified (Tektronix AM 502, OR, USA), band-pass filtered at 0.5–
6 kHz (Krohn Hite 3700, MA, USA), monitored for size and shape
on an oscilloscope, discriminated from noise by a window discrimin-
ator (WPI 121, CT, USA), digitized, displayed on-line as dot rasters
1304 R. Yaka et al.

FIG. 3. PSTHs (10 repetitions, bin duration –20 ms) of visual responses to a moving light bar recorded from 18 different cells in normal and BD cats. The bars
represent the preferred orientation and arrows represent the direction of the moving light bar. Lack of bars and arrows indicates no preferred orientation. Note
that visual responses of ALLS and AMLS cells are less orientation and direction selective, and less time-locked compared to those of A17.

and/or peristimulus time histograms (PSTHs), and stored for off-line
analyses on a PC. To reconstruct recording sites, reference electrical
lesions (20 µA, 10 s, anodal current) were performed at the end of
the recording session. Typically, a single recording session lasted 48–
96 h, at the end of which the animal was deeply anaesthetized with
2–4 mL of pentobarbitone sodium (60 mg/mL) and perfused (0.9%
saline/heparin followed by 10% neutral-buffered formalin). The brain
was then removed from the skull and kept for a few more days in
10% formalin. Cresyl violet-stained coronal sections (frozen or
paraffin) were used for the assignment of recording sites.

© 1999 European Neuroscience Association, European Journal of Neuroscience, 11, 1301–1312

 Auditory activation of cortical visual areas 1305

FIG. 4. Maps of visual receptive fields from ALLS, AMLS and A17 of a BD cat. LOD and ROD are the optical projections of the left and right optic disks,
respectively. LAC and RAC are the representation of the projections of the left and right area centralae, respectively. LM and RM are the theoretically determined
left and right vertical meridians of each visual field for each eye. Each receptive field is characterized by a cell number. R (right) and L (left), show the eye for
which each receptive field was mapped. I (ipsilateral) and C (contralateral) show the side of the eye for which the receptive fields were mapped with respect to
the hemisphere studied. Note the remarkable differences between the sizes of the ALLS and AMLS receptive fields and that of A17. Dotted lines indicate
boundaries of receptive fields that were classified as diffuse.

Visual receptive fields of isolated single cells were delineated
manually by projecting moving and stationary light bars and spots
on a tangent screen placed 2 m in front of the cat. Stimuli were
produced by a hand-driven projector with a light slit of changeable
dimensions. Following the manual session, a computerized optical
stimulation system was employed, automatically displaying the above-
described stimuli at various angles, directions, velocities and durations.
Auditory stimuli consisted of 0.1 ms clicks generated by a square-
pulse stimulator (Grass S88, MA, USA), broad-band (0.5–20 kHz)
white noise bursts (Wavetek 132, CA, USA), pure tones bursts (0.1–
35 kHz) produced by a PC-modulated voltage-controlled oscillator
(Tektronix FG 501) and presented in sequential steps of 0.2–1 kHz,
and frequency-modulated (FM) tones (0.1–35 kHz) generated by the
same voltage-controlled oscillator and modulated by a symmetrical
triangular waveform. The noise, pure tones and FM tones were shaped
into 200 ms bursts by a trapezoidal waveform with 15 ms rise and
fall time. Complex sounds, e.g. hand clapping, kissing and jingling
keys, were also used. Stimuli were attenuated, amplified (custom-
made attenuator and amplifier) and delivered binaurally through
calibrated earphones (JVC 144, Japan). The latter were located at the
outer apertures of the hollow ear bars and attached to them by means
of a short plastic tube. Calibration of the sound delivery system
was accomplished by simulating the experimental conditions. The
earphones were attached to the ear bars as described above, and a
condenser microphone (Brüel and Kjaer type 4134, Noerum,
Denmark) was facing the other end of the hollow bar and attached
to it by means of a short plastic speculum. The microphone was
connected to a sound level meter (type 2209) and a 1/3 octave filter
from the same company.
Cells that fired spontaneously and were held long enough were
tested with the entire auditory and visual stimulus repertoire. However,
in order to also detect responding cells that do not fire spontaneously,
we routinely used auditory and visual ‘searching’ stimuli (mainly
clicks, hand clapping, jingling keys, moving light bars and light
spots). Cells detected in this way were then tested with the entire
auditory and visual stimulus repertoire. Only cells that were tested
with the entire repertoire, including the complex sounds, were included
in this study.
Statistical data analysis
The difference between normal, BD and BE cats in terms of
responsiveness to visual and auditory stimuli was analysed using χ2
test for independent samples. In normal and BD cats, we compared
the relative numbers of auditory, visual, bimodal and non-responding
cells. As in BE cats there are no visual or bimodal cells, we had to
run a separate analysis comparing only the relative numbers of
auditory and non-responding cells. In this case, all the cells in the
normal and BD cats were divided into two groups, one consisting of
cells responding to auditory stimuli (exclusively auditory and bimodal
cells) and the other consisting of non-auditory cells (visual and non-
responding cells). In the BE cats there were either auditory or
non-responding cells. In the first series we also performed a pairwise
comparison within each modality group, including the non-responding
cells as a group (e.g. auditory cells in normal cats versus auditory
cells in BD cats, etc.). This was accomplished by applying the
Bonferroni correction for multiple comparisons. The effect of depriva-
tion on orientation and direction selectivity of visually driven cells

© 1999 European Neuroscience Association, European Journal of Neuroscience, 11, 1301–1312

1306 R. Yaka et al.

FIG. 5. Dot rasters (20 repetitions, 1 s apart) and their corresponding PSTHs (1 ms bin duration) illustrating responses of cells in the ALLS and AMLS of
normal, BD and BE cats to various auditory stimuli. NB, noise bursts; TB, pure tone bursts (presented at the best frequency of the cell); FM, frequency-
modulated tones swept symmetrically from 0.1 to 20.0 kHz and backwards; CL, clicks. The timing of the NB and TB is designated by horizontal bars at the
bottom of the dot rasters; that of the FM by the triangle modulating waveform and clicks by the arrows. Note the diversity of response patterns; the loose time-
locking of many of them with the stimulus; and their instability, manifested best in the dot rasters.

© 1999 European Neuroscience Association, European Journal of Neuroscience, 11, 1301–1312

 Auditory activation of cortical visual areas 1307

was evaluated statistically by a ‘z’ test (with Yates correction) for

comparing different proportions. The same test was also applied to
evaluate the effect of deprivation on response stability and response
time lock to auditory stimuli.

Results
General overview
The most apparent morphological effect of deprivation was the
complete degeneration of the optic nerve and optic chiasm in the BE
cats (Fig. 2). In BD cats these structures looked normal. Gross
morphology of the cortex in both BD and BE cats also looked normal,
and Nissl stain did not show any obvious cytoarchitectural changes
in any of the three examined areas.
In all three groups of cats the three areas A17, ALLS and AMLS
were surveyed for single-cell responses elicited by visual and auditory
stimuli. Cells that responded to at least one auditory stimulus out of
our entire repertoire were defined as auditory cells; those that
responded to at least one visual stimulus were termed visual cells.
Cells responding to at least one auditory and one visual stimuli were
defined as bimodal. Non-responding cells (NRs) were detected by
their spontaneous activity and lack of response to any of the auditory
or visual stimuli employed. Obviously, non-responding cells that did
not fire spontaneously could not be detected. Interestingly, in addition
to visual cells, auditory and bimodal cells were also found in the
ALLS and AMLS of normal cats. Because these areas are considered
to be strictly visual areas (Clare & Bishop, 1954; Hubel & Wiesel,
1969; Spear & Bauman, 1975; Palmer et al., 1978), this was an
unexpected finding.
Visual responses in the ALLS and AMLS in BD cats, as evaluated
subjectively by eye inspection, were less stimulus time-locked, less
stable, and less direction and orientation selective than those in
FIG. 6. Sensory modality distribution of cells in A17 (top panel), ALLS
normal cats (Fig. 3). This resulted in some difficulty in delineating (middle panel) and AMLS (bottom panel) of normal (NOR), eyelid sutured
the exact borders of visual receptive fields. The size of the receptive (BD) and bilaterally enucleated (BE) cats. The percentage of cells was
fields, both in the deprived and normal cats, was much larger than calculated from the total number of cells in each group of cats. vis, visual;
that typical of A17 cells; this was more noticeable in the AMLS (Fig. aud, auditory; bimod, bimodal; NR, no response. N 5 number of cats in each
group; n 5 number of cells in each group. Note in (A), the remarkable increase
4). Most of the responses to the various auditory stimuli in ALLS
in the relative number of non-responding cells as a result of deprivation; in
and AMLS, in all three cat groups, were also not very stable and not (B) and (C) that the increase in the relative number of auditory cells in the
highly stimulus time-locked (Fig. 5). Of the various auditory stimuli deprived cats becomes more prominent with increased severity of deprivation.
that we applied, pure tones were the least effective and tuning curves
were usually quite flat with relatively high thresholds (to values The ALLS area
known from the primary auditory cortex, e.g. Abeles & Goldstein,
1972), ranging between 40 and 70 dB (SPL re 20 µPa). It should be A total of 442 cells was studied in the ALLS of NOR, BD and BE
stressed, however, that this difference in thresholds could also be cats. The responsiveness of these cells to visual and auditory stimuli
attributed, at least partially, to differences in experimental procedures, is summarized in Fig. 6. It can be seen that in normal cats, 39% of
e.g. in the sound delivery system and anaesthesia. the cells responded only to visual stimuli, 14% were exclusively
Quantitative analyses regarding the effects of deprivation on some auditory, 1% were bimodal, and 47% failed to respond to either
response properties of visually and auditory-driven cells are described auditory or visual stimuli. Because the border between the auditory
in more detail in the following sections. cortex and the ALLS could not be rigorously determined, we can not
rule out the possibility that some of the auditory cells that were
located close to the border line were in fact primary auditory
Responsiveness to visual and auditory stimuli
cortex cells.
Area 17 In BD cats (Fig. 6) there was a significant decrease in the proportion
Figure 6 summarizes the responsiveness of the 278 A17 cells to of visual cells as compared to normal cats (χ2 5 58.39, P , 0.001)
visual and auditory stimuli, in the three groups of cats. It can be seen with a comparable increase in the relative number of auditory-
that in normal cats only visual stimuli were effective, with a high responding cells (χ2 5 45.02, P , 0.001). Most of these auditory
rate (87%) of responsiveness. In BD cats, as in normal cats, only cells were located deep in the lateral bank of the SSS, an area which
visual stimuli were effective, but responsiveness dropped slightly, was mostly visual in normal cats. The proportion of bimodal and
manifested by an increase in the relative number of non-responding non-responsive cells remained almost unchanged in the deprived cats.
cells. As expected, no visually driven cells were found in BE cats, In the BE cats (Fig. 6) there was an even more prominent increase
indicating that enucleation was indeed total. A small percentage of in the relative number of auditory cells (χ2 5 88.11, P , 0.001).
cells (6%) were driven by auditory stimuli. Percentage-wise, this increase was essentially equal to the total

© 1999 European Neuroscience Association, European Journal of Neuroscience, 11, 1301–1312

1308 R. Yaka et al.

FIG. 7. Distribution of cells types as a function of depth from the cortical surface in ALLS and AMLS of normal, BD and BE cats. Top: a coronal Nissl-stained
section through the anterior suprasylvian region (10 µm, A13 Horsley–Clarke coordinate) showing both the ALLS and AMLS. The scale, along the ALLS,
represents the depth in mm. The white arrow points towards an electrolytic lesion, and dorsally to it a longitudinal mechanical lesion, caused by the penetrating
electrode, is also discernible. The histograms represent the distribution of all ALLS (right) and AMLS (left) cell types, in all three animal groups, according to
their depth from the cortical surface, at a resolution of 1 mm. The total number of cells at each depth is indicated by the numbers above the bars for each
group. Distribution of the different cell types is presented for each depth as the percentage of the total number of cells at this depth. SSS, suprasylvian sulcus;
LS, lateral sulcus.

© 1999 European Neuroscience Association, European Journal of Neuroscience, 11, 1301–1312

 Auditory activation of cortical visual areas 1309

FIG. 8. Response properties of ALLS and AMLS cells to visual and auditory stimuli in normal and visually deprived cats. (A,B) Responses of normal and BD
cats to visual stimuli. (C,D) Responses of normal, BD and BE cats to auditory stimuli. n, number of cells; OS, orientation selective; OB, orientation bias; ONS,
orientation non-selective; DS, direction selective; DB, direction bias; DNS, direction non-selective; DIF, diffused receptive field; IC, incomplete receptive field.
Asterisks indicate significance levels in comparison to normal (NOR) cats using the ‘z’ test. *P , 0.05; **P , 0.01.

amount of visual and bimodal cells found in the normal cats plus the
decrease in the relative number of non-responding cells. The ‘overall
difference’ between normal, BD and BE cats was highly significant
(χ2 5 90.95, P , 0.001). Breaking down the overall distribution of
the different cell types in the ALLS in the three groups of cats,
according to their depth from the dorsal border, provided some
additional insight into the reorganization of these areas in the deprived
animals (Fig. 7). It can be seen that although the number of sampled
cells at the various depths (at a resolution of 1 mm), within and
between the three animal groups, was not identical, it was large
enough to make bias unlikely in our sampling. It can also be seen
that the gradual increase in the relative number of auditory cells as
impairment became more severe was not restricted to a particular
depth but occurred throughout the entire screened range depths.
The AMLS area
In the AMLS area a total of 434 cells was studied in normal, BD
and BE cats. The responsiveness of these cells to visual and auditory

© 1999 European Neuroscience Association, European Journal of Neuroscience, 11, 1301–1312

1310 R. Yaka et al.
stimuli is summarized in Fig. 6. It can be seen that most of the
responsive cells in normal cats were exclusively visual. However, an
appreciable amount of auditory and bimodal cells was also found.
About half of the cells did not respond to any of the stimuli used.
As in the ALLS, the most noticeable effect of deprivation on the
AMLS cells was an increase in the relative number of auditory-
responsive cells. This increase became more prominent as deprivation
became more severe, from 16% in normal cats to 28% in BD cats
(χ2 5 6.48, P , 0.01) and 64.5% in BE cats (χ2 5 93.15, P , 0.01).
In this area in BD cats, unlike the ALLS of similarly deprived
cats, there was only a slight and non-significant decrease in the
relative number of visual cells. The proportion of bimodal cells
dropped significantly (χ2 5 12.53, P , 0.001) and that of non-
responsive cells remained almost unchanged. In BE cats obviously
no visual cells were detected, and the increase in the relative number
of auditory cells was about the same as the total percentage of visual
and bimodal cells in normal cats in addition to the decrease in the
number of non-responding cells. The ‘overall difference’ between
normal, BD and BE cats was highly significant (χ2 5 17.29,
P , 0.001).
Regarding the distribution of the different cell types according to
their depth from the dorsal border (Fig. 7), it is apparent that, similar
to the ALLS, there was no bias for a particular depth in our sample
of cells, and that the conclusion we reached from the total sample is
essentially valid with regard to the different depths.
Response properties
As mentioned above, visual responses in the ALLS and AMLS in
normal and BD cats, as evaluated subjectively, were less stable and
less stimulus time-locked than those in normal cats. However, as
these responses were stored for off-line analyses only as cumulative
PSTHs of consecutive trials (Fig. 3), quantifying these differences
was essentially unfeasible. A more prominent difference between
normal and BD cats was manifested in direction and orientation
selectivity. Based on previous studies (Blakemore & Van Sluyters,
1974; Fregnac & Imbert, 1978), visually driven cells were classified
into three major subgroups according to: (i) ‘orientation selectivity’;
(ii) ‘direction selectivity’; and (iii) characteristics of ‘receptive field’.
Each of these three groups was further subdivided as follows. Group
‘a’ consisted of ‘orientation selective’ (OS) cells that responded to a
defined sector of angles and had an optimal stimulus orientation
within this sector; ‘orientation-bias’ (OB) cells that responded to the
entire range of stimulus orientations with an optimal orientation; and
‘orientation non-selective’ (ONS) cells that did not show any prefer-
able orientation. Group ‘b’ consisted of ‘direction selective’ (DS)
cells that responded only to one direction of stimulus movement (a
slit of light oriented at the receptive field axis and moving perpendic-
ularly to it); ‘direction bias’ (DB) cells that responded to both
directions but with some preference for one direction; and ‘direction
non-selective’ (DNS) cells that responded equally to both directions.
Group ‘c’ consisted of cells with ‘diffused’ (DIF) receptive fields
characterized by having indistinct border lines; and cells with ‘incom-
plete’ (IC) receptive fields for which one or mostly two of its borders
could not be determined.
Figure 8 summarizes the relative incidence of these various cells
among the total number of visually driven cells in normal and BD
cats. It can be seen that there is an obvious trend (though not
confirmed statistically in all cases, probably due to the relatively
small number of visually driven cells in BD cats) of ALLS and
AMLS cells in BD cats to become less orientation and direction
selective. This is manifested by a larger proportion of exclusive and/
or biased ‘orientation’ and ‘direction’ selective cells in normal cats,
© 1999 European Neuroscience Association, European Journal of Neuroscience, 11, 1301–1312
and a larger proportion of ‘non-selective’ cells in BD cats. In the
AMLS of BD cats, as compared with normal cats, there is also a
significantly larger proportion of cells with ‘incomplete’ receptive
fields.
Stability and stimulus time-lock of responses to auditory stimuli
were evaluated by comparing the responses to the same repeated
stimuli. A response was considered ‘stable’ when at least 75% of the
repeated stimuli elicited a response, regardless of fluctuations in
latencies. A response was considered as ‘stimulus time-locked’ when
response latencies to at least 75% of the repeated stimuli fluctuated
within 5 ms for short clicks and 15 ms for all other auditory stimuli.
In Fig. 8C and D it can be seen that in the ALLS of BD and BE cats
response stability and time-lock dropped significantly. In the AMLS,
a significantly reduced stability and stimulus time-locking was noticed
only in BE cats.
Discussion
In this study we looked for possible auditory activation of cortical
visual areas in cats that were visually deprived by suturing their
eyelids (BD cats) or impaired by bilateral enucleation (BE cats)
shortly after birth. The rationale for using BE cats in addition to BD
cats for studying neuroplasticity in visually impaired animals is based
on the fact that eyelid suturing does not cause complete deprivation
of light and thus does not represent the most severe visual impairment.
Unlike previous studies in which rerouting of one sensory modality
into brain areas of a deprived sensory modality was enhanced by also
ablating its original targets (Sur et al., 1990; Frost, 1990; Pallas &
Sur, 1993), we left the entire auditory pathway intact. We believe
that our approach more accurately simulates true pathological and
accidental cases of blindness in humans.
The areas that we investigated were the primary visual cortex (area
17) and two extrastriate cortical visual areas – the anterolateral lateral
suprasylvian (ALLS) and anteromedial lateral suprasylvian (AMLS).
Selection of these two extrastriate areas was made on the assumption
that their close proximity to the primary auditory cortex renders them
good candidates for that purpose. This working assumption was also
supported by the fact that in cats that were neonatally deprived of
vision by suturing their eyelids (BD cats), the anterior ectosylvian
visual cortex (AEV), another extrastriate visual area, was activated
by auditory and somatosensory input (Rauschecker & Korte, 1993).
Indeed, we found that in BD cats the ALLS and AMLS were activated
by auditory input. However, in neonatally enucleated (BE) cats there
was a much more prominent increase in the relative number of
auditory cells. Moreover, while in BD cats we did not find any
auditory-responding cells in area 17, we did find a small number of
auditory-driven cells in BE cats. It is thus evident that the more
severe the visual impairment, the more prominent is the auditory
manifestation. We also found that in normal cats the ALLS and
AMLS, areas considered as strictly visual (Clare & Bishop, 1954;
Hubel & Wiesel, 1969; Spear & Bauman, 1975; Palmer et al., 1978),
received auditory input, which was remarkably enhanced in the
impaired animals. This finding supports the suggestion of Rauschecker
& Korte (1993), that brain structures that receive input from more
than one sensory modality have a better chance of undergoing cross-
modal compensatory changes.
As yet, we have no indication as to the pathway and mechanism
that underlie this neuronal reorganization. Moreover, the question of
which cells turn into auditory cells is still open. However, based on
our results, it is tempting to suggest that most of the original visual
cells in BD cats, and all of them in BE cats, turned into auditory
cells. In BE cats, all the visual and bimodal cells, as well as

an appreciable number of non-responsive cells, went through this
transformation. Similarly, it is reasonable to assume that in the AMLS
of BD cats it was mainly the visual input to the bimodal cells that
was eliminated, leaving only the auditory input. In BE cats, all the
visual and bimodal cells, as well as a noticeable fraction of non-
responsive cells, turned into auditory cells. This may reflect, e.g. the
formation of new synapses, unmasking of latent inputs or a combina-
tion of the two.
This assumption, based on formal calculations, is one possible
interpretation of our results. However, other possibilities can not be
ruled out. If, e.g. in BE cats a defined fraction of the non-responsive
and all of the exclusively visual responders cease to fire spontaneously,
all the visual cells will become non-responding cells and the bimodal
cells will lose their visual response, one would end up with proportions
of non-responders and auditory cells similar to those described here.
If this later model is valid, it will mean that essentially the visual
cells did not in fact turn into auditory cells and the relative increase
in auditory cells merely reflects a situation in which many cells
simply died or turned off as a result of the visual deprivation.
Studies with congenitally anophthalmic mice, naturally blind mole
rats (Spalax ehrenbergi), the insectivorous mole (Mogera) and enuc-
leated hamsters have revealed an extensive presentation of somatosen-
sory and/or auditory input in primary visual targets (Bronchti et al.,
1989; Heil et al., 1991; Asanuma & Stanfield, 1990; Necker et al.,
1992; Toldi et al., 1994; Doron & Wollberg, 1994; Kudo et al., 1997;
Izraeli & Wollberg, 1998). It is pertinent to mention, with regard to
our findings in the blind mole rat, that other investigators have argued
that our identification of the dLGN was incorrect (Cooper et al.,
1993a, b). We believe, however, that the additional evidence we
provided in a later paper and the fallacies that we found in some of
their arguments (Doron & Wollberg, 1994) corroborate our original
assumption. Moreover, this assumption is also supported by the recent
finding that the common mole Mogera, an almost totally blind
insectivorous mammal, also possesses auditory projections from the
inferior colliculus to the dLGN (Kudo et al., 1997).
The activation of primary visual areas by other sensory modalities
found in blind animal models is in accordance with recent findings
from blind human subjects, as revealed by the use of various new
non-invasive brain-imaging techniques (Veraart et al., 1990; Alho
et al., 1993; Kujala et al., 1995; Sadato et al., 1996). We thus expected
a more substantial auditory activation of area 17 than that which we
observed in our animals, at least in BE cats. To our surprise we did
not find any auditory activity in area 17 of BD cats and only a very
small number of auditory cells in area 17 of BE cats. This, however,
supported earlier studies which also failed to elicit auditory or
somatosensory responses in the primary visual cortex of BD cats
(Rauschecker & Korte, 1993). The lack of auditory activity in area
17 of visually impaired cats is also in accordance with some earlier
projection tracing studies. It has been shown that the transitory
auditory projections to visual areas 17 and 18 found in kittens
degenerate almost completely during maturation. This happens both
in normal cats and in cats that have been bilaterally enucleated at
birth, suggesting that postnatal elimination of the auditory projections
to the visual cortex is independent of visual experience or the activity
of retinofugal projections (Innocenti et al., 1988). In some congenitally
blind or neonatally enucleated rodents (Bronchti et al., 1989; Heil
et al., 1991), and in blind human subjects (Veraart et al., 1990; Alho
et al., 1993; Kujala et al., 1995; Sadato et al., 1996), visual targets
have been found to be activated by other sensory modalities. Moreover,
in two subterranean mammals that are essentially blind, a rodent
(Doron & Wollberg, 1994) and an insectivore (Kudo et al., 1997),
the inferior colliculus also projects to the lateral geniculate nucleus.
© 1999 European Neuroscience Association, European Journal of Neuroscience, 11, 1301–1312
Auditory activation of cortical visual areas 1311
These findings raise the possibility that the patterns of neuronal cross-
modal reorganization are also species dependent.
Acknowledgements
We would like to thank I. Spivak for his help in the experiments, Dr G. Dobin
for help with the histology, E. Zimerman for technical assistance, and Ms N.
Paz for her help in preparing the manuscript. We also wish to thank Y. Parmet
for his assistance in the statistical analyses. We are most grateful to Dr R.
Izraeli for her critical review of the manuscript and Dr D. Bonimovich for
writing some of the computer programs. Last, but not least, we want to thank
the anonymous referees for their very constructive comments and useful
suggestions.
Abbreviations
A17, primary visual cortex; AEV, anterior ectosylvian visual cortex; ALLS,
anterolateral lateral suprasylvian area; AMLS, anteromedial lateral supra-
sylvian area; BD, binocularly deprived cats; BE, binocularly enucleated cats;
DB, direction bias cell; DIF, diffuse receptive field; dLGN, dorsal lateral
geniculate nucleus; DNS, direction non-selective cell; DS, direction selective
cell; IC, incomplete receptive field; LS, lateral sulcus; OB, orientation bias
cell; ONS, orientation non-selective cell; OS, orientation selective cell; PLS,
post-lateral sulcus; PSTH, peristimulus time histogram; SC, superior colliculus;
SSS, suprasylvian sulcus.
References
Abeles, M. & Goldstein, M.H. Jr (1972) Responses of single units in the
primary auditory cortex of the cat to tones and tone pairs. Brain Res., 42,
337–352.
Alho, K., Kujala, T., Paavilainen, P., Summala, H. & Naatanen, R. (1993)
Auditory processing in visual brain areas of the early blind: evidence
from event-related potentials. Electroencephalogr. Clin. Neurophysiol., 86,
418–427.
Asanuma, C. & Stanfield, B.B. (1990) Induction of somatic sensory inputs to
the lateral geniculate nucleus in congenitally blind mice and in
phenotypically normal mice. Neuroscience, 39, 533–545.
Bear, M.F., Cooper, L.N. & Ebner, F.F. (1987) A physiological basis for a
theory of synapse modification. Science, 237, 42–48.
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. (Lond.), 237, 195–216.
Bronchti, G., Heil, P., Scheich, H. & Wollberg, Z. (1989) Auditory pathway
and auditory activation of primary visual targets in the blind mole rat
(Spalax ehrenbergi), I. A 2-deoxyglucose study of subcortical centers.
J. Comp. Neurol., 284, 253–274.
Clare, M.S. & Bishop, G.H. (1954) Responses from an association area
secondarily activated from optic cortex. J. Neurophysiol., 21, 271–277.
Cooper, H.M., Herbin, M. & Nevo, E. (1993a) Ocular regression conceals
adaptive progression of the visual system in a blind subterranean mammal.
Nature, 361, 156–159.
Cooper, H.M., Herbin, M. & Nevo, E. (1993b) Visual system of a naturally
microphthalmic mammal: the blind mole rat, Spalax ehrenbergi. J. Comp.
Neurol., 328, 313–350.
De Volder, A.G., Bol, A., Blin, J., Robert, A., Arno, P., Grandin, C., Michel, C.
& Veraart, C. (1997) Brain energy metabolism in early blind subjects,
Neural activity in the visual cortex. Brain Res., 750, 235–244.
Doron, N. & Wollberg, Z. (1994) Cross-modal neuroplasticity in the blind
mole rat Spalax ehrenbergi, a WGA-HRP tracing study. NeuroReport, 5,
2697–2701.
Fregnac, Y. & Imbert, M. (1978) Early development of visual cortical
cells in normal and dark-reared kittens: relationship between orientation
selectivity and ocular dominance. J. Physiol. (Lond.), 278, 27–44.
Fregnac, Y., Shulz, D., Thorpe, S. & Beinenstok, E. (1992) Cellular analogs
of visual cortical epigenesis. I. Plasticity of orientation selectivity.
J. Neurosci., 12, 1280–1300.
Frost, D.O. (1990) Sensory processing by novel, experimentally induced
cross-modal circuits. Ann. NY Acad. Sci., 608, 92–112.
Heil, P., Bronchti, G., Wollberg, Z. & Scheich, H. (1991) Invasion of visual
cortex by the auditory system in the naturally blind mole rat. NeuroReport,
2, 735–738.
1312 R. Yaka et al.
Hubel, D.H. & Wiesel, T.N. (1969) Visual area of the lateral suprasylvian
gyrus (Clare-Bishop area) of the cat. J. Physiol. (Lond.), 202, 251–260.
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.
Innocenti, G.M., Berbel, P. & Clarke, S. (1988) Development of projections
from auditory to visual areas in the cat. J. Comp. Neurol., 272, 242–259.
Imig, T.J. & Reale, R.A. (1980) Patterns of cortico-cortical connections related
to the tonotopic maps in cat auditory cortex. J. Comp. Neurol., 192, 293–332.
Izraeli, R. & Wollberg, Z. (1998) Auditory takeover of the visual cortex in
visually deprived hamsters. Eur. J. Neurosci. Suppl., 10, 278.
Kaas, J.H. (1991) Plasticity of sensory and motor maps in adult mammals.
Ann. Rev. Neurosci., 14, 137–167.
Kasamatsu, T. & Adey, W.R. (1974) Recovery in visual cortical neurons
following total visual deafferentation. Brain Res., 74, 105–117.
Kellogg, W.N. (1962) Sonar system of the blind. Science, 137, 399–404.
Korte, M. & Rauschecker, J.P. (1993) Auditory spatial tuning of cortical
neurons is sharpened in cats with early blindness. J. Neurophysiol., 70,
1717–1721.
Kudo, M., Moriya, M. & Mizuno, N. (1997) Auditory projections from the
IC to the SCN by way of the LG in the mole, Mogera. Neuroreport, 8,
3405–3409.
Kujala, T., Huotilainen, M., Sinkkonen, J., Ahonen, A.I., Alho, K., Hamalainen,
M.S., Ilmoniemi, R.J., Kajola, M., Knuutila, J.E.T., Lavikainen, J.,
Salonen, O., Simola, J., Standertskjld-Nordenstam, C.-G., Tiitinen, H.,
Tissari, S.O. & Naatanen, R. (1995) Visual cortex activation in blind
humans during sound discrimination. Neurosci. Lett., 183, 143–146.
Marshall, W.H., Talbot, S.A. & Ades, H.W. (1943) Cortical responses of
the anesthetized cat to gross photic and electrical afferent stimulation.
J. Neurophysiol., 6, 1–15.
Merzenich, M.M., Kaas, J.H., Wall, J.T., Sur, M., Nelson, R.J. & Felleman,
D.J. (1983) Progression of change following median nerve section in the
cortical representation of the hand in areas 3b and 1 in adult owl and
squirrel monkeys. Neuroscience, 10, 639–665.
Necker, R., Rehkamper, G. & Nevo, E. (1992) Electrophysiological mapping
of body representation in the cortex of the blind mole rat. Neuroreport, 3,
505–508.
Niemeyer, W. & Starlinger, I. (1981) Do the blind hear better? investigations
on auditory processing in congenital or early acquired blindness. II. Central
functions. Audiology, 20, 510–515.
Otsuka, R. & Hassler, R. (1962) Uber aufbau und gleiderung der corticalen
sehsphare bei der katze. Arch. Psych. Zeitschr. Neurol., 203, 212–234.
Pallas, S.L. & Sur, M. (1993) Visual projections induced into the auditory
pathway of ferrets, II. Corticocortical connections of primary auditory
cortex. J. Comp. Neurol., 337, 317–333.
Palmer, L.A., Rosenquist, A.C. & Tusa, R.J. (1978) The retinotopic
organization of the lateral suprasylvian areas in the cat. J. Comp. Neurol.,
217, 237–256.
© 1999 European Neuroscience Association, European Journal of Neuroscience, 11, 1301–1312
Rauschecker, J.P. & Harris, L.R. (1983) Auditory compensation of the effects
of visual deprivation in the cat’s superior colliculus. Exp. Brain Res., 50,
69–83.
Rauschecker, J.P. & Korte, M. (1993) Auditory compensation for early
blindness in cat cerebral cortex. J. Neuroscience, 13, 4538–4548.
Rauschecker, J.P. & Kniepert, U. (1994) Auditory localization behavior in
visually deprived cats. Eur. J. Neurosci., 6, 149–160.
Sadato, N., Pascual-Leone, A., Grafman, J., Ibanez, V., Deiber, M.-P., Dold, G.
& Hallett, M. (1996) Activation of the primary visual cortex by Braille
reading in blind subjects. Nature, 380, 526–528.
Scannell, J.W., Blakemore, C. & Young, M.P. (1995) Analysis of connectivity
in the cat cerebral cortex. J. Neurosci., 15, 1463–1483.
Spear, P.D. & Baumann, T.P. (1975) Receptive fields characteristics of single
neurons in lateral suprasylvian visual area of the cat. J. Neurophysiol., 38,
1403–1420.
Spear, P.D., Tong, L. & Sawyer, C. (1983) Effects of binocular deprivation
on responses of cells in cat’s lateral suprasylvian visual cortex.
J. Neurophysiol., 49, 366–382.
Sur, M., Pallas, S.L. & Roe, A.W. (1990) Cross-modal plasticity in cortical
development, differentiation and specification of sensory neocortex. TINS,
13(6), 227–233.
Toldi, J., Farkas, T. & Völgy, B. (1994) Neonatal enucleation induces cross-
modal changes in the barrel cortex of rat. A behavioral and
electrophysiological study. Neurosci. Lett., 167, 1–4.
Turlejski, K. & Michalski, A. (1975) Clare-Bishop area in the cat, Location
and retinotopic organization. Acta Neurobiol. Exp., 35, 219–228.
Uhl, F., Franzen, P., Lindinger, G., Lang, W. & Deecke, L. (1991) On the
functionality of the visually deprived occipital cortex in early blind persons.
Neurosci. Lett., 124, 256–259.
Veraart, C., De Volder, A.G., Wanet-Defalque, M.C., Bol, A., Michel, C. &
Goffinet, A.M. (1990) Glucose utilization in human visual cortex is
abnormally elevated in blindness of early onset but decreased in blindness
of late onset. Brain Res., 510, 115–121.
Woods, D.L., Clayworth, C.C. & Boch-Rita, P. (1985) Early blindness
reorganizes auditory processing in humans. Soc. Neurosci. Abstr., 11, 49.
Woolsey, C.N. (1979) Comparative studies on cortical representation of vision.
Vision Res., 3, 365–382.
Yaka, R., Yinon, U. & Wollberg, Z. (1996) Auditory takeover of extrastriate
visual areas in early binocularly deprived cats, physiological evidence. Soc.
Neurosci. Abstr., 22, 1062.
Yinon, U. (1980) Monocular deafferentation effects on responsiveness of
cortical cells in adult cats. Brain Res., 199, 299–306.
Yinon, U., Shemesh, R., Arda, H., Dobin, G. & Jaross, P.P. (1993) Physiological
studies in deafferented visual cortex cells of cats following transplantation
of fetal xeongrafts from the rat’s cortex. Exp. Neurol., 122, 335–341.
Yinon, U. & Goshen, S. (1984) Ocular dominance of cortical cells in cats
dark reared into maturity after short postnatal monocular deprivation. Exp.
Neurol., 85, 461–468.