Receptive-field properties of neurons in different laminae of visual cortex of the cat.

JOURNALOF NEUROPHYSIOLOGY Vol. 41, No. 4, July 1978. Printed

in U.S.A.

Receptive-Field Properties of Neurons in Different Laminae of Visual Cortex of the Cat AUDIE

GENE

LEVENTHAL

Center for Neurobiology,

SUMMARY

AND

AND

HELMUT

State Universiry

of New

CONCLUSIONS

I. Receptive-field properties of neurons in the different layers of the visual cortex of normal adult cats were analyzed quantitatively. Neurons were classified into one of two groups: I) S-cells, which have discrete on- and/or off-regions in their receptive fields and possess inhibitory side bands; 2) C-cells, which do not have discrete on- and off-regions in their receptive fields but display an on-off response to flashing stimuli. Neurons of this type rarely display sideband inhibition. 2. As a group, S-cells display lower relative degrees of binocularity and are more selective for stimulus orientation than Ccells. In addition, within a given lamina the S-cells have smaller receptive fields, lower cutoff velocities, lower peak responses to visual stimulation, and lower spontaneous activity than do the C-cells. 3. S-cells in all layers of the cortex display similar orientation sensitivities, mean spontaneous discharge rates, peak response to visual stimulation, and degrees of binocularity. 4. Many of the receptive-field properties of cortical cells vary with laminar location. Receptive-field sizes and cutoff velocities of S-cells and of C-cells are greater in layers V and VI than in layers II-IV. For S-cells, preferred velocities are also greater in layers V and VI than in layers II-IV. Furthermore, C-cells in layers V and VI display high mean spontaneous discharge rates, weak orientation preferences, high relative degrees of binocularity, and higher peak responses to visual stimulation when compared to C-cells in layers II and III. Received 948

for publication

December

19, 1977.

0022-3077/78/0000-0000$01.25

V. B. HIRSCH York, Albany,

New

York 12222

5. The receptive-field properties of cells in layers V-VI of the striate cortex suggest that most neurons that have their somata in these laminae receive afferents from LGNd Y-cells. Hence, our results suggest that afferents from LGNd Y-cells may play a major part in the cortical control of subcortical visual functions. INTRODUCTION

To understand how the visual system processes sensory information, experimenters have, in several species, assayed the receptive-field properties of neurons in the cortical regions that receive visual afferents (l-2, 4-7, 16, 20, 25, 33, 3639). Visual cortical neurons have been obseved to respond selectively to a number of stimulus parameters, such as orientation, direction, velocity, and size. These response properties vary among neurons that subserve different regions of the visual field (22, 48), as well as among neurons in the various visual cortical areas (24, 26,42). A number of studies indicate that the receptive-field characteristics of a cortical neuron are also related to the lamina within which its responses can be recorded (16, 22, 25, 37). In particular, layer IV of the striate cortex has been reported to contain a high proportion of simple cells, while layers V and VI have been observed to contain mainly complex cells (16, 22). A concentration of hypercomplex cells has been observed in cortical laminae II and III (10, 25). The aim of this study is to relate a number of the properties of receptive fields to the different cortical laminae of area 17 of the normal cat. Also, since many studies of the normal mammalian visual cortex have utiCopyright

0 1978 The

American

Physiological

Society

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 9, 2018 Copyright © 1978 the American Physiological Society. All rights reserved.

RESPONSES

OF CORTICAL

lized the (simple/complex/hypercomplex) classification proposed by Hubel and Wiesel (22), the choice was made to employ a similar system of classification in order to allow comparisons between the present results and those obtained by others. The results of this study indicate that, in normal cats, differences exist among many of the receptive-field characteristics of neurons sampled from the different cortical laminae. This is true even within each of the classes of cortical cells studied. In particular, cells sampled from the various cortical laminae exhibit different receptivefield sizes, velocity selectivities, orientation selectivities, degrees of binocularity, spontaneous discharge rates, and peak response rates, METHODS

Physiological recording procedures Seven normal adult cats were prepared individually for electrophysiological recording in a conventional manner (22, 31, 33). Either Fluothane or Pentothal sodium was used to induce anesthesia. Each animal was positioned in a stereotaxic apparatus and its skull was affixed to a rigid bar with screws and dental cement. The skin, bone, and dura mater covering the striate cortex were then removed and a cylindrical chamber was positioned over this craniotomy. The chamber was filled with a 4% solution of agar in saline to limit pulsations of the brain and to protect the cortex. All pressure points and incisions were infiltrated with a long-acting local anesthetic (Anucain). A mixture of dtubocurarine (0.7 mglkg* h) and gallamine triethiodide (7 mg/kg lh) was infused intravenously and the animal was ventilated with a mixture of nitrous oxide (80%), oxygen (19%), and carbon dioxide (1%). Body temperature was maintained at 38°C and heart rate was monitored throughout the experiment. Neo-Synephrine was used to retract the nictating membrane and atropine was used to dilate the pupils. The eyes were protected from dessication with contact lenses and, when necessary, spectacle lenses were used to focus the eyes on a tangent screen positioned 57 cm from the cat. The projections of the optic disks were determined repeatedly during the course of each recording session and were used to infer the position of the area centralis (8). In some instances the location of the area centralis was determined directly, and this location did not differ significantly from that inferred from the projections of the optic disks.

CELLS

IN LAYERS

II-VI

949

Insl-X-coated tungsten microelectrodes were employed to record the action potentials of cortical cells (22). For some penetrations the electrode was directed over the cortex so that it would pass through the cortical projection of the area centralis. In other penetrations more peripheral regions were examined. A hydraulic drive advanced the electrode obliquely through the cortex. The electrode was advanced at least 75 pm between units to reduce sampling bias and to record from many different columns of orientation-sensitive cells (2, 23, 46). Responses of units were amplified in a conventional manner and single units were isolated using a waveshape analyzer (40) l

Characterization

of response pruperties

RECEPTIVE-FIELDSIZE. Thereceptivefield of a cortical neuron was defined as the largest area in visual space within which a visual stimulus elicited a response. For most cells, the “minimum response field” was plotted using light bars and both light and dark edges in a manner similar to that described by Barlow et al. (3). A detailed description of our procedure has been given previously (3 1). RECEPTIVE-FIELD ECCENTRICITY. The eccentricity of a cell’s receptive field was defined as the distance from the center of the receptive field (determined by presenting stimuli to the dominant eye) to the projection of the area centralis for that eye. For all units studied, the most recent determination of the projection of the optic disks was used to infer the location of the area centralis. ORIENTATION SELECTIVITY. Orientation tuning curves were obtained utilizing an optical display capable of presenting moving, barshaped stimuli at various orientations, directions, and velocities. This display was used in conjunction with a digital counter, which monitored the number of action potentials elicited by each stimulus presentation. All stimuli were presented to the eye which elicited the strongest response from the unit under investigation. The direction of stimulus motion was always in one of the two directions orthogonal to the orientation of the stimulus being presented; the particular direction employed was the one that was determined initially to be the more effective. Similarly, the velocity employed was also the one judged to be optimal. Typically, 5-15 stimulus presentations at each of 8-15 orientations were used to compile an orientation tuning curve. For comparison, the animal’s eyes were occluded and the mean spontaneous discharge rate was determined under comparable conditions. Two measures of orientation selectivity were employed in this study. They are 1) width of tuning, and 2) half-width at half-height. The


950

A. G. LEVENTHAL

AND

width of tuning of a cortical neuron is defined as the range of stimulus orientations over which it responds. Half-width at half-height is obtained from two intersecting regression lines which are fitted to the orientation tuning curve (Figs. 4, 5) and is defined as one-half of the width of that tuning curve taken halfway between the peak of the curve (the preferred orientation of the cell) and the maintained discharge rate of the unit. VELOCITY SELECTIVITY. Velocity tuning curves were obtained by determining the mean number of impulsesthat were evoked by appropriately oriented and directed stimuli moved at various velocities through the cell’s receptive field. Typically, two to five presentations of a stimulusmoving at velocities rangingfrom 1.5 to 17O”lswere employed. In this fashion, the cutoff velocity (the maximal stimulus velocity to which a cell would respond) and the preferred velocity (the velocity to which a unit gave its strongestresponse)were determinedfor all cells studied. OCULAR DOMINANCE. A quantitative measure of the effectiveness of the two eyes in influencing the response rate of each neuron was determinedby presentingappropriately oriented stimuli, moving at the cell’s preferred velocity, to each eye separately. The responseevoked by 5-10 stimulus presentations to each eye was then determinedand the meanresponseper presentation was calculated. A ratio, ocular dominance ratio, or OD, indicating the relative effectiveness of the dominant and of the nondominant eye was computed after adjusting for the unit’s spontaneousdischarge rate: (R (dominant eye) - SA)/(R (nondominanteye) - SA), where SA is the spontaneous activity and R is the mean responseper presentation. To insure that small differences between the preferred orientation of the cell determined through each eye separately would not be confounded with differences in ocular dominance, the ocular dominance ratios of most cells were determined three times utilizing three different stimulus orientations. Cells studied in this fashion were tested with stimuli oriented at the preferred orientation for the unit (determined through the dominant eye), as well as with stimuli oriented 10’ clockwise and 10’ counterclockwise from the orientation preference of the cell. For most cells, the preferred orientation that had been determined through the dominant eye was also most effective for the other, nondominant eye. The ocular dominance ratio of cells for which this was not the casewas computed, utilizing the most effective orientation for eacheye. SPONTANEOUS ACTIVITY. The spontaneous dischargesof each cortical neuron were counted

H. V, B. HIRSCH

for 30 s and the mean spontaneousdischarge rate per secondwas calculated. PEAK RESPONSE. The peak response that could be evoked visually was determined by computing the meannumber of action potentials evoked by three to five presentations of an optimally oriented and directed stimulus presented to the dominant eye of each unit. In all cases, the stimulus was moved at the velocity that evoked the greatestnumber of action potentials from the cell. SIDE-BAND INHIBITION. The excitatory discharge center and inhibitory sidebandsof a simple cell are mutually antagonistic (4-6). Thus, simultaneousstimulation of both regions should evoke a response that is smaller than the responseevoked if the excitatory region alone is stimulated. In order to test for the presenceof side-bandinhibition, a distinguishingcharacteristic of cortical simple cells (5-7), a sampleof cells was tested both with optimal stimuli alone (20-40 trials presentedin groups of 5) and with optimal stimuli presentedwhile a light bar (1” x 5’) was being flashed on and off parallel to the receptive field of the cell.

Classification

of units

Neurons were placed into one of two groups basedon their responsesto flashing stimuli and the presence or absence of distinct inhibitory subdivisionsin their responsefields. The classification systemwas meant to be similarto the one proposedby Hubel and Wiesel(22). As hasbeen suggestedby Schiller and co-workers (36), however, it is likely that different experimenters employ slightly different criteria to categorize cortical cells. In addition, the terminology suggested cannot be easily reconciled with any theory of cortical organization other than the hierarchical schemethat Hubel and Wiesel proposed(18). For thesereasons,the two classesof cortical neurons studied here are referred to simply as S-cells and C-cells. The receptive fields of S-cells can be subdivided into discrete on- and/or off-regionsbased on their responseto flashingstimuli. Many cells of this type display side-band inhibition. The receptive fields of C-cells cannot be subdivided into discrete on- and off-regions and C-cells give an on-off responseor no responseto flashing stimuli. Few cells of this type exhibit inhibitory side bands, Finally, units that displayed “endstop inhibition” were classifiedas S-cells or Ccells, depending on their response to flashing stimuli (16).

His tology Electrolytic lesions (5-10 PA for 5-10 s; electrode negative) were madeat intervals along


RESPONSES

OF CORTICAL

all electrode tracks. In some animals, smaller electrolytic lesions (3 PA for 3 s) were made at the site of recording of cells of particular interest. These smaller lesions, marking the location of individual cells, could be differentiated from the larger ones that defined the penetration itself. At the termination of the electrophysiological recording session, all animals were given an overdose of anesthetic and then perfused with physiological saline followed by a mixture of 10% formalin in saline. The areas of cortex containing the electrode tracks were then embedded in paraffin and sectioned at 8-10 pm. Alternate slides (each containing lo-20 sections) were stained with cresyl violet or 1~x01 fast blue. In this fashion, electrode tracks could be reconstructed and recording sites could be localized to the different laminae within area 17. A penetration was considered to have been reconstructed adequately only if: I) its beginning and end were located unambiguously, 2) its length (after allowing for shrinkage) corresponded to the microdrive readings recorded during the experiment, and 3) a number of lesions (usually one per millimeter) located along the track were localized. Units which could not be assigned a laminar location were considered only in the comparison of the total populations of S-cells and C-cells.

Statistics The probabilities that the differences observed would have occurred by chance are given within the text. These probabilities are not based on conventional inferential statistics, since many of the assumptions of standard statistics (e.g., independence of observations) cannot be met for these data. Rather, each probability was based on 500 computer simulations. During each simulation, the observed data were randomly permuted and assigned to the categories such that a cat which provided IZ observed data values to a particular category (e.g., S-cells in layer IV) would also provide n values to that category during the simulation. In addition, the data were randomized only across the variable being tested; classification with respect to other variables was held constant. Thus, to compare S-cells with Ccells overall, the data were randomized with respect to receptive-field type, while classification with respect to depth in cortex remained fixed. Likewise, to compare overall differences between cortical layers, the data were randomized with respect to laminae, while classification with respect to S-cell or C-cell remained fixed. The descriptive statistics that were calculated on the observed data were also calculated on the results obtained from each appropriate computer simulation. For example, when comparing

CELLS

IN LAYERS

II-VI

951

pairs of means (e.g., receptive-field sizes of Scells and C-cells), an appropriate descriptive statistic would be the difference between the means However, when comparing several means (e.g., receptive-field sizes of cells in the different layers), an appropriate statistic would be the variance of the means. The probabilities given in the text indicate the proportion of 500 simulations in which the absolute value of the descriptive statistic computed for the simulation was greater than the absolute value of the corresponding statistic for the data as they were actually obtained in the experiments. A low probability indicates that the observed difference or variance of the means was not likely to have occurred by chance. A high probability indicates very little difference between or among the means. Generalizations to other samples of cells or to other samples of cats must be made with the customary caution. Details of this method of statistical analysis are presented elsewhere (ID. G. Tieman, unpublished data). l

RESULTS

The response characteristics of 158 neurons sampled from seven normal adult cats were assessed. The laminar position of 132 of these cells was determined and these neurons were placed in one of three groups based on their depth in the cortex. The first group consisted of cells judged to be in either layers II or III, the second consisted of cells in layer IV, and the third consisted of cells in either layers V or VI. Since the positions of only a limited number of cells were marked by individual lesions, it was felt that a further subdivision of the cortex was not justified. Receptive-field area The present results indicate that receptive-field size varies with depth in the cortex and that differences in receptive-field area are evident within as well as between the two classes of cells included in the study. In particular, S-cells, in general, have smaller receptive fields than do C-cells (P = 0.00). In addition, as can be seen in the histograms for receptive-field size, the receptive fields of S-cells (Fig. lB, left) are smaller than those of C-cells (Fig. lB, right) in every layer. Finally, the receptive fields of both Sand C-cells are larger for neurons in layers V and VI than for neurons in either lavers


A. G. LEVENTHAL

952

A

I

II-m

Ip

i

,

II and III (P = 0.00) or in layer IV (P = 0.08). The receptive-field sizes of cells sampled during one oblique microelectrode penetration through area 17 are presented in Fig. 1A. The units studied during this penetration illustrate the differences in the receptive-field sizes of S- and C-cells, as well as the increase in receptive-field size of both Sand C-cells in the deeper layers of the cortex.

x-91

1 I

Scetts 0 C cells l

0

1

0 0

I

I

I

AND H. V. B. HIRSCH

0 0

Velocity selectivity VELLOCJTY. The results of this study indicate that cells found in layers V and VI have higher cutoff velocities than do cells in either layers II and III (P = 0.00) or in layer IV (P = 0.00). This is true for both S-cells (P = 0.01) and C-cells (P = 0.03), although C-cells (Fig. 2B, right) generally have higher cutoff velocities than do S-cells (Fig. 2B, left) (P = 0.01). The cutoff velocities of neurons sampled during one long electrode penetration are shown in Fig. 2A. Differences between S-cells and Ccells are evident, as is the increase in cutoff velocity of both S-cells and C-cells in the deeper layers of the cortex. PREFERRED VELOCITY. The present results suggest that a relatively large proportion of the cells in layers V and VI of the visual cortex prefer rapid stimulus motion. This appeared true for both S-cells and Ccells. Overall, C-cells (Fig. 3, right) did not prefer higher stimulus velocities than did Scells (Fig. 3, left) (P = 0.60). A considerable number of S-cells in layers V and VI had higher preferred velocities than did some of the C-cells in layers II and III. The distribution of velocity preferences of cortical cells found in the various laminae is shown in Fig. 3. Clearly, the differences in preferred velocity which could be attributed to laminar differences were much greater within the class of S-cells (P = 0.00) than within the class of C-cells (P = 0.98). Velocity tuning curves typical of S-cells and of C-cells sampled from the different laminae are illustrated in Figs. 4B and 5B, respectively. For each unit shown, the lower curve indicates the total number of action potentials elicited by stimuli moving at each velocity tested. The upper curve repreCUTOFF

*aa

a0

0

1

l6m Distance

I

2400

3200

(microns)

L

i = 6,8 (N=16)

La

HI -~~ ’ c ii 1=2,8 (N=X))

=10,5

b z &

(N=14)

I 80 60

E = 6.0 (N=21)

Receptive

w & VI

f iekl ore0 ( deprees’

I = I?,? IN=461

1

FIG. 1. A: receptive-field areas of neurons studied during a long, oblique microelectrode penetration through layers II-VI of the striate cortex. Approximate borders between laminae are indicated by dashed lines, Solid symbols indicate S-type cells, open symbols indicate C-type cells, and squares indicate units with considerable end-stop inhibition. All cells included in this figure had receptive fields within 3.4’ of the area centralis. B: receptive-field areas of S-type cells (left) and of C-type cells (right) sampled from the different cortical laminae. The number of cells represented in each histogram as well as the mean for each distribution are indicated.

ownloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (130.063.180.147) on December 9, 201 Copyright © 1978 the American Physiological Society. All rights reserved.

RESPONSES

OF CORTICAL

CELLS

IN LAYERS

II-VI

953

C

S 160 140 -

j

\8

l20-

loo-

0 0

)I .= 80P z 3 60-

I

I

’ 0

I

I

I I

I I 00II I

0 0 I ’ 0 0: 0

I0

20-

’

i

m*I

I

0 00

1

I

I I

a

I l6a

0

Distom

B

50 40 30 x) IO 0 50

000 0

x = IL7 (N=lO)

60

h

90

0

60

60

i

l

Scells

0

Ccells

I 2400

1 ‘3200

60 45

1

t

60

c

f = 37. I (N=l8)

wI H

z y

40 30

2

$50 28

i= ?8,4 (N=t4)

40

Preferred

velocity

60 45

E = 4515 (N=39)

30 I5

(degrees/second

1

FIG. 3. Preferred velocities of S-type cells (left) and of C-type cells (right) sampled from the different cortical laminae. The number of cells represented in each histogram as well as the mean for each distribution are indicated.

30 20

IO

0 ‘ii = 128.2 (N=46)

(N=l4)

30 h

i=lo6.5 (N=85)

g20 8 to : 0

E = 33.3 (N=l5)

30 15

c

S ‘iit61.4 (N=58)

X = 3813 (N=76)

45

I I I 1 hicms

c

90

I I l

l

0 0

45 30

I

0

40-

60

I I I

g = 2415 (N=59)

r

velccity ( FIG. 2. A: cutoff velocities of neurons studied during a long, oblique microelectrode penetration through layers II-VI of the striate cortex. Approximate borders between laminae are indicated by dashed lines. Solid symbols indicate S-type cells, open symbols indicate C-type cells, and squares indicate units with considerable end-stop inhibition. All cells included in this figure had receptive fields within 3-4’ of the area centralis. These data are from the same penetration as that in Fig. 1A. B: cutoff velocities of S-type cells (left) and of C-type cells (right) sampled from the different cortical laminae. The number of cells represented in each histogram as well as the mean for each distribution are indicated.

sents the response of the cell adjusted for the length of time that the stimulus remained within the response field of the neuron. The preferred velocity of each unit can be estimated from the position of the peak of the upper curve with respect to the X axis. Examination of the velocity tuning curves indicates that there are both S- and C-type cells in layers V and VI with velocity tuning curves that differ markedly from the velocity tuning curves characteristic of either Scells or C-cells in layers II and III. The response of these layer V-VI cells does not fall off as rapidly with increasing stimulus velocity as does the response of most S-cells and C-cells found in the upper portions of the cortex. S-cells displaying curves of the type


A. G. LEVENTHAL

ORIENTATION -

TUNING A

AND

H. V. B. HIRSCH

S CELLS

CURVES

VELOCITY

3or

IOI-

TUNING

B

CURVES

15or

v) 4. .-cl E

'g 20 .-s 2

0

0 g

8-

i-?

0 i2 Orientation

in degrees

Degrees

8-

g-

per second

FIG. 4. A: computer-generated orientation tuning curves of S-type cortical neurons studied in normal cats. Data points have been fitted by two regression lines whose intersection determines the preferred orientation of the cell. Regression lines have been computed utilizing all points greater than the mean spontaneous discharge rate of the cell: plus 10% of the response evoked by the most effective orientation tested. For all units, the width at halfheight (twice the half-width) as well as the mean spontaneous discharge rate are indicated by lines parallel to the X axis. Top: cells in layers II- III; middle: cells in layer IV; bottom: cells in layers V-VI. B: computer-generated velocity tuning curves of S-type neurons studied in normal cats. Original data points (action potentials/stimuius presentation) are indicated on the lower curve. These data points are adjusted for the time that the stimulus remained within the response field of the cell and are displayed on the upper curve. Upper curves are either unsmoothed (data points indicated by filled circles) or are smoothed using a running box-car average of each two adjacent points (means indicated by open circles) or of each three adjacent points (means indicated by XS). Top: cells in layers II- III; middle: cells in layer IV; bottom: cells in layers V-VI.

shown in Fig. 4B, bottom row? and C-cells exhibiting curves similar to those displayed in Fig. 5B, bottom row, were rarely found in the upper layers of the cortex. Orien t&ion s4ktivity S-cells (Fig. 6, left) were observed to be more sensitive to stimulus orientation than C-cells (Fig. 6, right) (P = 0.02). This difference was most evident in the lower cortical laminae, since the orientation selectivity of C-cells (P = 0.15), but not of S-cells (P = 0.62). varied appreciably with depth in the cortex. That is, C-cells studied in layers II and III were more selective for stimulus orientation (mean half-width at half-height, 24.5”) than were C-cells in the lower laminae (mean half-width at half-height, 33.5”) (Fig. 6). This trend wits especially marked when the proportion (67%) of C-cells in layers V

and VI that responded to all orientations (tuning widths of 180”) was compared to the proportion (47%) of C-cells of this type in the upper laminae (Fig. 6). In layers V and VI there is thus a pronounced increase in the number of C-cells that give some response to a line presented at any orientation. Examples of orientation tuning curves typical of S-cells and of C-cells studied in the different cortical laminae are shown in Figs. 4A and 5A, respectively. These illustrate the relative uniformity across the different cortical laminae of S-cell orientation selectivity, as well as the reduced orientation selectivity characteristic of some of the Ccells found in the deeper cortical laminae. Oculur dominunce Typically, S-cells having respo nse fields within 5” of t he proj ection of the are a cen-


RESPONSES

ORIENTATION

TUNING

OF CORTICAL

CURVES

CELLS

IN

C CELLS

LAYERS

VELOCITY

TUNING

CURVES

B

A 36

955

II-VI

I8

E

E

36?

R A

I35

E

63 42c

1

63 42

E-

21 0

36

1

270

Orientation

F

in degrees

FIG. 5. A : computer-generated orientation tuning curves of C-type neurons studied in normal cats. Data points have been fitted by two regression lines whose intersection determines the preferred orientation of the cell. Regression lines have been computed utilizing all points greater than the mean spontaneous discharge rate of the cell plus 10% of the response evoked by the most effective orientation tested. For all units, the width at half-height (twice the half-width) as well as the mean spontaneous discharge rate are indicated by lines parallel to theX axis. Top: cells in layers II-III; middle: cells in layer IV; bottom: cells in layers V-VI. B: computer-generated velocity tuning curves of C-type neurons studied in normal cats. Original data points (action potentials/stimulus presentation) are indicated on the lower curve. These data points are adjusted for the time that the stimulus remained within the response field of the cell and are displayed on the upper curve. Upper curves are either unsmoothed (data points indicated by solid circles) or are smoothed using a running box-car average of each two adjacent points (means indicated by open circles) or of each three adjacent points (means indicated by Top: cells in layers II-III; middle: cells in layer IV; bottom: cells in layers V-VI.

XS).

tralis were less likely t han we re C-ce 11sto be activated comparably by the two ey es (Fig74 w (P = 0.08). In particular, only 39% of the S-cells but 63% of the C-cells studied near the cortical projection of the area centralis were activated well by either eye (ocular dominance ratios less than 3.0). Such a low relative degree of binocular activation was evident among S-cells in all of the cortical laminae (P = 0.61). In contrast, differences in ocular dominance that were related to laminar location were observed within the class of C-cells (P = 0.14). Specifically, while 76% of the C-cells in layers V-VI were activated well by either eye (ocular dominance ratio less than 3.0), only 45% of the C-cells in layers II and III responded well to stimulation of either eye. Peak response

Cells located in the lower layers of the visual cortex generally responded to appro-

priate visual stimulation more strongly than did cells in the upper laminae (Fig. 8) (P = 0.02). While laminar differences within the group of S-cells were not obvious (P = 0.57) (Fig. 8B, left), there was some increase in the peak response of C-cells in the deeper laminae (P = 0.19) (Fig. 8B, right). It was not unusual for certain C-cells in layers VVI of the striate cortex to respond to optimal stimuli with long, rather erratic bursts of 60- 150 action potentials. C-cells of this type were rarely observed in the upper layers of the visual cortex. Spontaneous activity

The spontaneous discharge rates of Scells and C-cells studied in layers II-VI are shown in Fig. 9. S-cells display relatively low rates of spontaneous activity compared to C-cells (P = 0.00). In general, the spontaneous discharge rates of cells in layers V and VI are higher than the spon-


956

A. G. LEVENTHAL

si = 30.4

I

40 30 20 IO 0

AND H. V. B. HIRSCH

(N=79) l 0

Scells Ccells

mm* I

l

D .e ‘OcI3 L 83 6l l

4-

I 0

c

E = 33,5 (N=4t)

1 I

s

r

1 I t600 Distonee I microns)

Eccentricity

N=48

I 2400

I 3200

O-5

C

c

N=50

60 40 20 0

Holf

width

(degrees

1

FIG. 6, Orientation selectivity (half-width at halfheight) of S-type cells (left) and of C-type cells (right) sampled from the different cortical laminae. The number of cells represented in each histogram as well as the mean for each distribution are indicated. Inset histograms indicate the proportion of neurons in each distribution that displayed at least some response to all stimulus orientations (width, 180”) as well as cells that responded to only a limited range of stimulus orientation (width less than or equal to 179”).

taneous discharge rates of cells in layers II and III (P = 0.01). Most of this effect is due to the C-cells (P = O.Ol), since the spontaneous activity of the S-cells does not change with depth in cortex (P =0.45). End-stup inhibition For a limited number of neurons, the elongation of an optimally oriented stimulus beyond the borders of the discharge region resulted in a reduced response. Most neurons with this property displayed responses that were otherwise typical of S-type cells and were found in the upper cortical laminae (Figs. lA, 2A, 7A, 9A). These cells appear similar to type I hypercomplex cells (13) and may simply be a variation of S-type or simple cells (34).

80 c

N=lO

fi P 60 40 20 0

80

L

N =I9

N=7

60

m 6

dominance

0 -T

6

ti

(rutio)

FIG. 7. A: ocular dominance ratios of neurons studied during a long, oblique microelectrode penetration through layers II-VI of the striate cortex. Approximate borders between laminae are indicated by dashed lines. Solid symbols indicate S-type cells, open symbols indicate C-type cells, and squares indicate units with considerable end-stop inhibition. All cells included in this figure had receptive fields within 3.4’ of the area centralis. These data are from the same penetration as that in Fig. 1A. B: ocular dominance ratios of S-type cells (left) and of C-type cells (right) sampled from the different cortical laminae. The number of cells represented in each histogram is indicated.


RESPONSES

OF CORTICAL

CELLS

IN LAYERS

lI-III

A -

Ip

;

IO g-

l

Scells

I

$7-

0 Ccells

I I I

i

957

II-VI P-xl

I

0

I I I I

0

0

0

0

0

I

20

0

t

1

l

IO

lo

0

t a

o” mm

0

l

0’

f

I 1 i 800

l *

I I

l*

l

O

0

w

l

1 1 Im Distance (micrax

I 2400

1 3200

ii(N=lO) =9*5

z

,-,

i = 23.2 (N=l5)

E (action potentials

/ ‘stimulus

I I **..d

FIG. 8. A: peak response to visual stimulation of neurons studied during a long, oblique microelectrode penetration through layers II-VI of the striate cortex. Approximate borders between laminae are indicated by dashed lines. Solid symbols indicate S-type cells, open symbols indicate C-type cells, and squares indicate units with considerable end-stop inhibition. All cells included in this figure had receptive fields within 3.4” of the area centralis. These data are from the same penetration as that in Fig. 1A. B: peak response to visual stimulation of S-type cells (left) and of C-type cells (right) sampled from the different cortical laminae. The number of cells represented in each histogram as well as the mean for each distribution are indicated.

l

I

1

2400

3200

S

C i= I.7 (N=77)

F

40 30 20 IO 0 50 40 30 20

n

F

f = 45.6 (N=44)

presentation)

1

l6W Distance ( rnicmns)

50 IO 0

Peak response

l p?o

0

50 40 30 20 IQ 0

40 30 20 10 0

50

,253 l m 0 D o”

1

SO

40

I0

I

0

Spontonems

activity

( action pdentiols

q 2.4 (N=40) i

/second )

FIG. 9. A: mean spontaneous discharge rates of neurons studied during a long, oblique microelectrode penetration through layers II - VI of the striate cortex. Approximate borders between laminae are indicated by dashed lines. Solid symbols indicate S-type cells, open symbols indicate C-type cells, and squares indicate units with considerable end-stop inhibition All cells included in this figure had receptive fields within 3.4’ of the area centralis. These data are from the same penetration as that in Fig. IA. B: mean spontaneous discharge rates of S-type cells (left) and of C-type cells (right) sampled from the different cortical laminae. The number of cells represented in each histogram as well as the mean for each distribution are indicated.


958

Receptive-field eccentricity location of units studied

A. G. LEVENTHAL

and laminar

It has been observed that if a microelectrode enters area 17 near the cortical projection of the area centralis and is driven either obliquely through area 17 or down the medial bank of the postlateral gyrus, there is a tendency for the response fields of neurons sampled from layers V and VI to have greater eccentricities than those of cells sampled from the upper laminae. Presumably, this is because the electrode has traveled some distance from its starting point before it reaches the lower cortical layers. Since the receptive-field properties of cortical cells change both with laminar location (16, 22) and with eccentricity (48), care must be taken to insure that differences related to the eccentricity of the response fields being studied are not being confounded with differences related to the layers from which cells are sampled. In order to determine the extent to which this problem affected the present findings, the distribution of receptive-field eccentricities of cells included in this study were examined. The results of this analysis indicate that all units included in the present study had response fields within 16” of the area centralis and 88% of these cells had receptive fields within 10” of the area centralis. In addition, the mean eccentricity of the response fields of all S-cells in the present sample was 4.4’ (range, O-15.5’) and of all C-cells was 5.7” (range, 0.2-16”). Finally, it is noteworthy that the laminar differences evident in Figs. lA, 2A, 7A, 8A, and 9A indicate that differences in visualfield position do not pose a significant problem to the present study; all the units included in these figures had receptive fields within 3.4” of the area centralis. The probability levels given in the text were computed using only data from cells with receptive fields centered within 5’ of the area centralis. Hence, the statistics used present a very conservative estimate of the relationships present in the data and use only part of the data illustrated in each figure. Some of the illustrations may suggest stronger relationships than are implied by the statistics. However, even in the most extreme case (laminar changes in preferred velocity for C-cells, Fig. 3), the statistics

AND

H. V. B. HIRSCH

computed for all cells (P = 0.5 1) do not lead to a substantially different interpretation than is provided by the set of cells located within 5’ of the area centralis (P = 0.98). To further minimize any differences that could be attributed to eccentricity, a linear correction was applied for each variable. Thus, the variation that could be predicted from eccentricity was subtracted and the statistics repeated. In no case did this correction affect the pattern of results sufficiently to alter the conclusions. DISCUSSION

The results of this study indicate that Scells in a given layer of the striate cortex have smaller response fields, lower cutoff velocities, and lower peak responses than do C-cells in the same lamina. In addition, S-cells, regardless of their laminar location, display low relative degrees of binocular interaction and are quite selective for stimulus orientation compared to C-cells. The present findings also suggest that certain of the receptive-field properties of cortical cells vary with laminar location, while others apparently remain constant. Within the class of S-type cortical cells, receptive-field size, cutoff velocity, and preferred velocity increase with depth in the cortex (especially in layers V and VI). However, S-cells in all layers of the cortex display similar orientation selectivities, mean spontaneous discharge rates, degrees of binocular interaction, and peak firing rates. Laminar differences also exist within the class of C-cells. In particular, C-cells in the lower cortical laminae display high mean spontaneous discharge rates, weak orientation preferences, and high relative degrees of binocular interaction compared to Ccells in the upper layers. In addition, the receptive-field sizes, cutoff velocities, and peak firing rates of C-cells increase with depth (especially in layers V and VI) in the cortex. Relation

to previous

work

A number of the relationships reported here have been reported to exist within and between the classes of simple cells and complex cells in the striate cortex of the cat (16, 19, 22, 47, 48). Thus, it is likely that units included in this study have been classi-


RESPONSES

OF CORTICAL

fied in a manner similar to that proposed by Hubel and Wiesel (22). Gilbert (16) has also studied laminar differences in the response properties of neurons in the visual cortex of the cat. Since he divided complex cells into two groups, a direct comparison between Gilbert’s cell types and those described in the present study is difficult. Certain similarities between Gilbert’s results and ours are, however, evident. Gilbert also found that receptive-field sizes and spontaneous activity levels of cortical cells found in layers V and VI were higher than those of cells found in the upper layers of the cortex. Furthermore, a greater proportion of the cells in layers V and VI than in the upper cortical layers were activated in a comparable fashion by the two eyes. Finally, Gilbert also observed that simple cells showed a lower degree of binocularity than either of his two groups of complex cells. On the other hand, Gilbert reports that simple cells and standard complex cells have comparable tuning widths, velocity preferences, directional selectivity, and spontaneous activity levels, whereas we found that S-cells have lower preferred velocities, lower peak responses, and are more selective for stimulus orientation than C-cells. Since Gilbert’s second class of complex cells, special complex cells, had high-velocity preferences and high peak responses, overall differences between Gilbert’s simple and complex cells may not differ that much from the differences we observed between S- and C-cells. In fact, the remaining point of disagreement, the differences in tuning width between simple and complex cells, is difficult to evaluate, since Gilbert only determined the orientation selectivity of one-third of the cells in his sample. Certain of the laminar differences observed here for the cat have been reported by Schiller and co-workers (36-39) to exist in monkey. In particular, these authors have proposed a system of classification similar to the one employed in this study and report that S-type striate cortical neurons in all layers of the primate cortex have small receptive fields, low spontaneous discharge rates, and marked selectivity for stimulus orientation compared to CX-type cells. In

CELLS

IN LAYERS

H-VI

959

addition, CX-cells in layers II-III of the monkey’s visual cortex were reported to display small response fields, low rates of spontaneous activity, pronounced orientation sensitivity, and low relative degrees of binocular interaction when compared to CX-cells in the lower layers of the striate cortex. In summary, there is overall agreement about the nature of the laminar differences in the response properties of cells in the visual cortex, even though the classification of cortical cells remains a topic of controversy. Presumably, some of the differences within the class of C-cells reflect the presence in layers V-VI of cells that project either to the superior colliculus (32, 35, 41) or to the lateral geniculate nucleus (16, 17). These cells are reported to have very large receptive fields, preferences for high rates of stimulus motion, and high degrees of binocular activation when compared to other cortical cells. In cats deprived of early visual experience for extended periods, cells with small response fields and low cutoff velocities display apparently normal orientation preferences. Most of these cells are activated monocularly. In contrast, in these same cats, cells with large response fields and/or high cutoff velocities do not display orientation preferences but can be activated binocularly (31). The present findings indicate that the class of cells with small response fields and low cutoff velocities described previously (3 1) consists largely of cells that can be classified as simple, using a classification system similar to the one proposed by Hubel and Wiesel(22). However, it is likely that this class also contains some cells in layers II-III that can be termed complex. The remaining cells, those with large fields and/or high cutoff velocities, appear to be primarily cells that can be termed complex; however, a limited number of these cells in layers V-VI can be termed simple. Ajfiwnt cwznec*tivity Hoffmann and Stone (20) have suggested that cortical simple cells receive afferent inputs from LGNd X-cells, whereas many cortical complex cehs receive LGNd Y-cell afferents. In addition, it has been observed that cortical cells that are influenced by Xcells have small receptive fields and prefer-


A. G. LEVENTHAL

ences for low rates of stimulus motion, whereas neurons that are influenced by Ycells exhibit larger receptive fields and preferences for rapid stimulus motion (20, 43). The present results, thus, are compatible with the suggestion that many simple cells in the striate cortex receive LGNd X-cell afferents and that many complex cells in the striate cortex receive inputs from Y-cells in the LGNd. However, the small relative receptive-field sizes and the low relative cutoff velocities of some C-cells in layers IIIII, as well as the large relative receptivefield sizes and the high relative cutoff velocities of some S-cells in layers V-VI, are not reconciled very easily with this hypothesis. It is likely, thus, that many complex cells in layers II-III receive substantial Xcell or W-cell inputs and that some simple cells in cortical layers V-VI receive significant Y-cell or W-cell inputs. In fact, it has been suggested that some complex cells receive X-cell afferents (43) and some simple cells receive Y-cell afferents (28,41). The receptive-field sizes and preferred velocities of cortical neurons have been reported to increase with increasing eccentricity (22, 48). These differences may, in part, reflect the relative abundance of Y-cell inputs to eccentric cortical regions since retinal and LGNd Y-cells have large receptive fields and prefer rapid stimulus motion (11,21,44,45). The present findings indicate that the receptive-field sizes, cutoff velocities, and preferred velocities of cortical neurons are also greatest in layers V and VI. This indicates that the influence of LGNd Y-cells is stronger in the lower cortical layers while that of X-cells is stronger in the upper laminae. Conceivably, the projection from the lateral geniculate nucleus to layer VI of the cat’s visual cortex (30) consists mainly of the axons of Y-cells in the LGNd. Finally, the role of W-cells, as well as the cortical distribution of their afferents from the LGNd, remains an open question. Such an inhomogeneity in the cortical distribution of geniculate afferents may provide insight into the functional organization of the visual cortex. In particular, since corticofugal fibers originate mostly in layers V and VI of area 17, Gilbert (16) has suggested that the deeper layers of the visual cortex are involved in the cortical control of subcortical visual functions, notably eye

AND

H. V. B. HIRSCH

movements, and in feedback to the LGNd. Based on our results, we further suggest that it is primarily the cortical neurons dominated by afferents from LGNd Y-cells that modulate subcortical visual function. This suggestion is consistent with the finding that Y-cells in the retina and the thalamus are morphologically distinct and have relatively large cell bodies (9, IS), since the cells which project from area 17 to thalamus and tectum are relatively large pyramidal cells (17). Hence, the Y-cell afferent pathway in different parts of the visual system appears to involve cells which have similar morphological characteristics. In addition, we suggest that the relatively small cells, which are concentrated in the more superficial layers of the cortex (especially the stellate cells of layer IV), are dominated by afferents from LGNd X-cells. Clearly, these suggestions must be tested. They could, however, explain why some simple cells were identified morphologically as pyramidal cells while some complex cells were identified as stellate cells (29), since our results suggest that some simple cells receive Y-cell afferents while some complex cells receive X-cell afferents. It may, therefore, be possible to demonstrate stronger correlations between the receptive-field properties and the morphology of cortical cells if the cortical neurons are classified, at least partially, according to the nature of their afferent inputs. The relatively large number of Y-cells subserving the peripheral retina, their transient response, their large receptive fields, their responsiveness to a wide range of stimulus velocities, and their projection to both the superior colliculus and forebrain, suggest that these neurons mediate certain aspects of peripheral vision such as the detection of stimulus motion (11, 12, 14, 27, 44). On the other hand, the sustained response of X-cells, their concentration in the central retina, their small receptive fields, and their pattern of projection to the forebrain indicates that these neurons are suited for high-resolution vision (11, 12, 14, 27, 44). Differences in the characteristics of Xand Y-cells thus provide further evidence for the functional significance of the postulated laminar distribution of afferent input from the lateral geniculate nucleus to the visual cortex.


RESPONSES

OF CORTICAL

ACKNoWI+EDCrMENTS

We thank P. Caruccio and S. Johnson for their expert technical assistance. D. Greenberg participated in some of the experiments. L. Schmidt provided valuable advice regarding the examination of the histological material. B. Dreher, J. Stone. D. G. Tieman, and S. B, Tieman read the manuscript critically, J. Leventhal, R. Loos, T. Pacer, R. Speck, L. Stern, and L. Welch assisted greatly in the preparation of the manuscript.

CELLS

IN LAYERS

II-VI

961

Support for this study was provided by Public Health Service Research Grant R01 EY-01268 and Alfred P. Sloan Foundation Fellowship BR 1677.

Reprint requests should be addressed to H. V. B. Hirsch. Present address of A, G. Leventhal: School of Anatomy, University of New South Wales, P.O. BOX I, Kensington, N . S. W. Australia 2033.

REFERENCES 1. ALBUS,

K. A quantitative study of the projection of the central and the paracentrat visual field in area 17 of the cat. I. The precision of topography. E,vptl. Brain Rm 24: 1S9- 179. 1975. AI-BUS. K. A quantitative study of the projection area of the central and the paracentral visual field in area 17 of the cat. II. The spatial organization of the orientation domain. E+yptl. Brain &r-. 24: l8f-202, 1975. BARLOW, H. B., BLAKEMORE, C., AND PETTIGREW, J. D. The neural mechanism of binocular depth discrimination. J. Pl?~sicll. Lmdm 193: 327-342, 1967, BISHOP, P. 0.. C~OMBS, J. D., AND HENRY, G. H. Interaction effects of visual contours on the discharge frequency of simple striate neurones. J. Physiol. Londm 2 19: 659-687. 197 I. BISHOP, P. O., COOMBS, J. D.. AND HENRY, G. H. Responses to visual contours: spatio-temporal aspects of excitation in the receptive fields of simple striate neurones. J. Physisl. Lcmdcm 2 19: 625-657, 1971. BISHOP, P. 0.. COOMBS. J. D.. AND HENRY. G. H. Receptive fields of simple cells in the cat striate cortex. .I. Physiol. Lmdm 321: 3 l-60, 1973. BISHOP, P. O., HENRY, G. H., AND SMITH. C. J. Binocular interaction fields of single units in the cat striate cortex. J. Ph>Tsiol. Londo~~ 216: 3968, 1971. BISHOP, P. O., KOZAK, W., AND VAKKUR. G. J. Some quantitative aspects of the cat’s eye: axis and plane of reference, visual field co-ordinates and optics. J, Physiol. London 163: 466-502. 1962. BOYCOTT, B. B. AND WASSIX, H. The morphological types of ganglion cells of the domestic cat’s retina. J. PhysicA. London 240: 397-419, 1974. CAMARDA, R. AND RIZZOLATTI, G. Receptive fields of cells in the superficial layers of the cat’s area 17. Exptl. Bwin Rm. 24: 423-427, 1976. CLELAND, B. G.. DUBIN, M. W., AND LEVIC-K, W. R. Sustained and transient neurones in the cat’s retina and lateral geniculate nucleus. J. Physiol. London 217: 473-496, 1971. CLELAND, B. G., LEVIC-K, W. R., AND SANDEKSON, K. J, Properties of sustained and transient ganglion cells in the cat retina. ,/. Ph_vsicll. Lmdm 228: 649-680, 1973, DREHER, B. Hypercomplex cells in the cat’s striate cortex. Invest. Ophth&ml. 11: 355-356. 1972. ENROTH-CUGELL, C. AN~D ROBSON. J. G. The contrast sensitivity of retinal ganglion cells of the cat, J. Physiol. London 187: 517-552, 1966. area

2.

3,

4.

5.

6.

7.

8.

9

10.

11.

12.

13. 14.

15. FUKUDA, Y. AND STONE. J. Direct identification of the cell bodies of Y-. X- and W-cells in the cat’s retina. Vision Rcs. 15: 1034- 1036, 1975. 16. GILBERT. C. D. Laminar differences in receptive field properties of cells in cat primary visual cortex. ,I. Ph~siol. London 268: 391-421 . 1977. 17. GILBERT, C. D. AND KF.I.I.Y. J. P. The projection of cells in different layers of the cat’s visual cortex. J. Camp. Ncwrwl. 163: 8 I - 106, 1975. 18. HENRY, G. H. Receptive field classes of cells in the striate cortex of the cat. Bruin Rcs. 133: I28, 1977. 19. HENRY, G. H., DREHER, B., AND BISHOP, P. 0. Orientation specificity of cells in cat striate cortex. .I. Nuwophysiol. 37: 1394- 1409, 1974. 20. HOFFMANN, K.-P. AND STONE, J. Conduction velocity of afferents to cat visual cortex: a correlation with cortical receptive field properties. Brcrin RPS. 32: 460-466. 1971. 21. HOFFMANN, K.-P.. STONE. J.. AND SHERMAN, S. M. Relay of receptive-field properties in dorsal lateral geniculate nucleus of the cat. .I. NcwophysicA. 35: 5 18-53 I , 1472. 22. HUBEL, D. H. AND WII:SEI_. T. N. Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex. .I. Yhysiol. Lmdrm 160: 106- 154, 1962. 23. HUBEI,, D. H. AND WIESEL. T. N. Shape and arrangement of columns in cat’s striate cortex. J. Physiol. Londm 165: 559-568. 1963. D. H. AND WIESEL, T. N. Receptive 24. HUBEI,, fields and functional architecture in the two nonstriate visual areas ( 18 and 19) of the cat. J. Ncwwphysiol. 28: 229- 289. 1965. 25. HUBEL, D. H. AND WIESEL,. T. N. Receptive fields and functional architecture of monkey striate cortex. J. Physiol. Londorr 19s: 2 15 -243. 1968. D. H. AND WIESEI.. T. N. Visual area of 26. HUBEL, the lateral suprasylvian gyrus (Clare-Bishop area) of the cat. .I. Physiol. London 202: 25 I-260, 1969. 27. IKEDA, H. AND WRIGHT, M. J. Receptive field organization of ‘sustained’ and ‘transient’ retinal ganglion cells which subserve different functional roles. J. Physiol. London 227: 769-800. 1972. 28. IKEDA, H. AND WRIGHT. M. J. Retinotopic distribution, visual latency and orientation tuning of ‘sustained’ and ‘transient’ cortical neurones in area 17 of the cat. E.u/>tl. Burin Rrs. 22: 385398, 1975. 29. KELLY, J. P. AND VAN ESSIZN, D. C. Cell structure and function in the visual cortex of the cat. .I. Physiol. London 238: 5 M-547, 1974. 30. LEVAY, S. AND GMERT, C. 0. Laminar patterns


A. G. LEVENTHAL of geniculocortical projection in the cat. Brain Res. 113: I- 19, 1976. 31, LEVENTHAL, A. GANDHIRSCH, H.V.B.Effects of early experience upon orientation sensitivity and binocularity of neurons in visual cortex of cats. Prop. Natl. Acad. Sci, US 74: 1272-1276, 1977. 32. PALMER, L. A. AND R~SENQUIST, A. C. Visual receptive fields of single striate cortical units projecting to the superior colliculus in the cat. Brain Res. 67: 27-42, 1974. PETTIGREW, J. D., NIKARA, T,, AND BISHOP, I? 0. Responses to moving slits by single units in cat striate cortex. Exptl, Bruin Res. 6: 373-390, 1968, ROSE, D. The hypercomplex cell classification in the cat’s striate cortex. J. Physiol. London 242: l23P, 1974. ROSENQUIST, A. C. AND PALMER, L. A. Visual receptive field properties of ceils of the superior colliculus after cortical lesions in the cat. Exptl. N~urol. 33: 629-652, 1971. SCHILLER, P. H., FINLAY, B, L., AND VOLMAN, S. F. Quantitative studies of single-cell properties in monkey striate cortex. I. Spatiotemporal organization of receptive fields. J, Ncurophysiol. 39: 1288-1319, 1976. SCHILLER, P. H., FINLAY, B. L., AND VOLMAN, S. F. Quantitative studies of single-cell properties in monkey striate cortex. II, Orientation specificity and ocular dominance. J. Neurophysiol. 39: 1320-1333, 1976. SCHILLER, P. H,, FINLAY, B. L., AND VOLMAN, S. F. Quantitative studies of single-cell properties in monkey striate cortex. III. Spatial frequency. J. Neurophysiol. 39: 1334- 1351, 1976. SCHILLER, P. H., FINLAY, B. L., AND VOLMAN, S. F. Quantitative studies of single-cell properties

AND

40.

41.

42.

43,

44.

45.

46.

47.

48.

H. V. B. HIRSCH in monkey striate cortex. IV. Multivariate statistical analysis and models. J. Neurophysiol. 39: 1362-1374, 1976, SCHMIDT, E. M. An instrument for separation of multiple-unit neuroelectric signals. IEEE Trans. Bio-Med. Eng. 18: 155-157, 1971. SINGER, W., TRETTER, F., AND CYNADER, M. Organization of cat striate cortex: a correlation of receptive-field properties with afferent and efferent connections. J. Neurophysiol. 38: 10801098, 1975. SPEAR, P. D. AND BAUMANN, T. P. Receptivefield characteristics of single neurons in lateral suprasylvian visual area of the cat. J. Neurophysiol. 38: 1403-1420, 1975, STONE, J. AND DREHER, B, Projection ofXand Y-cells of the cat’s lateral geniculate nucleus to areas 17 and 18 of visual cortex. J. Neurophysiol. 36: 551-567, 1973. STONE J. AND FUKUDA, Y. Properties of cat retinal ganglion cells: a comparison of W-cells with X- and Y-cells. J. Neurophysiol. 37: 749-772, 1974. STONE, J. AND HOFFMANN, K.-P. Very slowconducting ganglion cells in the cat’s retina: a major, new functional type? Bruin Res. 43: 6 lo616, 1972. STRYKER, M. P. AND SHERK, H. Modification of cortical orientation selectivity in the cat by restricted visual experience: a re-examination. Science 190: 904-905, 1975. WATKINS, D, W. AND BERKLEY, M. A. The orientation selectivity of single neurons in cat striate cortex. Exptl. Brain Res, 19: 433-446, 1974. WILSON, J. R. AND SHERMAN, S. M. Receptive field characteristics of neurons in cat striate cortex: changes with visual field eccentricity+ J, Neurophysiol. 39: 512-533, 1976.


Maintained activity of single neurons in the cat visual cortex at different levels of retinal adaptation.

Mechanisms underlying the receptive field properties of neurons in cat visual cortex.

Morphology of physiologically identified neurons in the visual cortex of the cat.

Effect of Contrast on Visual Spatial Summation in Different Cell Categories in Cat Primary Visual Cortex.

Mechanisms of inhibition in cat visual cortex.

Visual response properties of neurons in the middle and lateral suprasylvian cortices of the behaving cat.

Visual properties of neurons in the posterior suprasylvian gyrus of the cat.

Organization of suppression in receptive fields of neurons in cat visual cortex.

Morphological types of projection neurons in layer 5 of cat visual cortex.

Local circuit neurons of macaque monkey striate cortex: III. Neurons of laminae 4B, 4A, and 3B.

18 border neurons contributing to the visual transcallosal pathway in the cat.

Modification of orientation sensitivity of cat visual cortex neurons by removal of GABA-mediated inhibition.

Siamese cat: altered connections of visual cortex.

Contrast invariance of orientation tuning in cat primary visual cortex neurons depends on stimulus size.

Laminar differences in receptive field properties of cells in cat primary visual cortex.

Spatial frequency filters in cat visual cortex?

Reward related neurons in cat association cortex.

Receptive fields and response properties of neurons in the rat visual cortex.

Distribution and properties of commissural and other neurons in cat sensorimotor cortex.

Centrifugal and centripetal connections of the cat visual cortex.

Binocular summation and the binocularity of cat visual cortex.

Optimal spatial displacement for direction selectivity in cat visual cortex neurons.

An intracellular analysis of the visual responses of neurones in cat visual cortex.

Quantitative analysis of unit response patterns in cat visual cortex.