CLARSBISHOP AREA IN THE CAT: LOCATION AIVD RETINOTOPICAL PROJECTION

ACTA NEUROBIOL. EXP. 1975, 35: 179488 CLARSBISHOP AREA IN THE CAT: LOCATION AIVD RETINOTOPICAL PROJECTION Krzysztof TURLEJSKI and Andrzej MICHALSKI Department of Neurophysiology, Nencki Institute of Experimental Biology Warsaw. Poland Abstract. Visual responses of single units in the cortex of the middle suprasylvian sulcus were evaluated in the pretrigeminal cat. Electrode penetrations which passed through the Clare-Bishop area were in 94Oio of the cases within the stereotaxic coordinates A2 and A8. Within the Clare-Bishop area 40 /o of the units were responsive to the visual stimuli employed whereas adjacent to it only 10-20 /o responded. Correlation between receptive field size and eccentricity was poor or absent. Most receptive field centers lay within the lower contralateral quadrant. Because of the large size of the receptive fields, their scatter and individual variability, it was not possible to depict a precise scheme of retinotopical projection. INTRODUCTION The Clare-Bishop area in the cat's cortex was discovered by Marshall et al. (13) in analysing photically evoked potentials. Midway along the medial portion of the middle suprasylvian sulcus the potentials were different from those in the surrounding association cortex, and similar to the primary responses of visual cortex. Later Clare and Bishop (2) found that the area is strongly influenced by the visual cortex. There are few data on the borders and retinotopy of the Clare-Bishop area. Evoked potentials placed the area around the center of the suprasylvian sulcus (2, 13, 17). Hubel and Wiesel (12), when recording single unit responses, penetrated between Horsley-Clarke coordinates A4 and A6 1 - Acta Neurobiologiae Experimentalis

180 K. TURLEJSKI and A. MICHALSKI that are close to the border of the caudal third of the sulcus. On the basis of the topography of anatomical connections, others believe that the Clare-Bishop area covers all the medial wall of the middle suprasylvian sulcus (6, 9, 15, 16). This is consonant with Palmer's analysis of multiunit discharge (14). The retinotopy of the area has been described by Hubel and Wiesel (12). They found that the receptive fields represented the contralateral visual field, and most were positioned below the horizontal meridian. They felt that the representation of the upper visual field occurred taudally to their penetrations. The peripheral visual field was represented at the top of the sulcus, while in the deeper regions receptive fields tended towards the area centralis. The receptive field positions were scattered considerably; thus the retinotopy was not very precise. Similarly, Palmer (14) found the representation of the contralateral visual field in the central part of the sulcus. Lower and to the sides the representation was central. According to Hubel and Wiesel (12) the receptive fields are large (sometimes one quadrant of the visual field) but they diminish for central fields. Wright (19) described small (2-7 deg) fields, but he recorded only from the bottom of the sulcus. We continued such examination of the site and retinotopy of Clare- Bishop area, since there is obviously some uncertainty extant concerning these important features. METHODS Experiments were conducted on 18 adult cats, weighing 2-4 kg. The data on the receptive field size and suprasylvian sulcus position were also taken from other experiments performed on 13 cats. Pretrigeminal transection was made according to the technique of Zernicki (20). The animal was then further immobilized with Flaxedil (5-7 mg/kg X 1 hr). Physiological support systems for the animal are described in a subsequent paper (18). Craniotomy was made over the full length of the medial suprasylvian sulcus and then the sulcus was photographed or mapped in Horsley-Clarke AP and lateral coordinates. Then, the blind spots of the eyes were mapped with a reversible ophthalmoscope. The screen was scaled in rectangular coordinates of inclination (y) and declination (6). The values of declination were 2 deg larger than in Bishop's (1) scale, because the nodal point of the coordinates was placed at the crossing of the vertical midline and the Horsley-Clarke horizontal zero plane.

CLARE-BISHOP AREA IN THE CAT 181 Glass microelectrodes filled with 3 M KC1 were introduced at an angle of 30 deg laterally from the vertical. The amplified signal from the electrode was fed to an oscilloscope and loudspeaker. Because the determination of as many receptive fields as possible in every experiment was necessary for the description of retinotopical projection, in most of the experiments the action potentials were not recorded. We simply evaluated whether or not the cell responded to the binocularly applied stimuli and then defined the receptive field. Receptive fields were drawn in the rectangular coordinates described above, and radial coordinates (deviation w and position angle 11-1) were established for the field tenter. In one experimental series the post-stimulus histograms (PSH) were ysed to check the evaluation of the field position. In most cases the disparities were insignificant. Stimuli were moved by hand. They were black paper rectangles and circles, a diffuse flash and a slit of light 10 deg long. The luminance of the screen was 2 X 10-2 cdjm2, and that of the stimuli 2-4 cdim2. Units that did not react to the applied stimuli were counted as nonvisual. The third group was formed of the cells that showed no maintained activity, having been destroyed during penetration. In every experiment we tried to make penetrations 0.5-1 mm apart along the Clare-Bishop area and to find its borders. A part of the penetration was classified as belonging to the Clare-Bishop area if: (a) at least 20 /o of the units responded to our stimuli, not less than five were reactive, and the sector of penetration was longer than 500 pm. (b) some of the conditions of (a) were not fulfilled, but when the map of unit positions was drawn, the units were close to the part of another penetration classified as the Clare-Bishop area. RESULTS Location of the Clare-Bishop area Ninety eight penetrations were made and 78 of them passed through the Clare-Bishop area (Fig. 1). There are only a few penetrations outside the area because the exploration was stopped when the border of the area was found. In most cases the visually responding cells were first encountered 1 mm below the surface of the cortex and were absent before the bottom of the sulcus was reached. Half of the Clare-Bishop penetrations were classified according to rule (b) (see Methods). Most of such penetrations would have been in the border areas, but in some experiments visual activity of the whole area was generally low. More

182 K. TURLEJSKI and A. MICHALSKI Fig. 1. Position of suprasylvian sulcus and Clare-Bishop area in Horsley-Clarke AP coordinates. Abscissa, AP coordinates. A, anterior end of the suprasylvian sulcus; B, medial point of the sulcus; C, border of the posterior one third of length of the sulcus; D, posterior end of the sulcus; E, penetrations where the Clare-Bi- :hop area was found; F, penetrations where the Clare-Bishop area was absent. Broken lines, median values of the groups; N, number of cases in the gro~up. than 40 /o of the cells in the Clare-Bishop area and 10-20 /o of the cells outside it responded to visual stimuli, but individual differences were considerable. Nonvisual and silent cells were nearly equally frequent. When all the penetrations had been classified according to Horsley- Clarke AP coordinate (Fig. I), it was found that 94O/a of all the penetrations passing through the Clare-Bishop area were positioned between the coordinates A2 and A8. In most cases this part of the suprasylvian sulcus contained the full length of the Clare-Bishop area. However, the number of unsuccessful penetrations in this sector was comparatively large (15O/a). When it was necessary to locate the Clare-Bishop area in one penetration, the point of entry was always chosen in relation to the position and length of the particular suprasylvian sulcus. We found that 95O/o of the penetrations were successful if placed between 113 and 112 of the length of the sulcus from its posterior end. However, 26O/o of the successful penetrations were found outside this segment, because it is shorter (mean 2.8 mm) than the Clare-Bishop area (mean 3.1 mm). The correlation of the successful penetrations with the center of the sulcus (plus and minus 2 mm) is not satisfactory.

CLARE-BISHOP AREA IN THE CAT 183 Size of the receptive fields and retinotopy The areas of 446 receptive fields were evaluated (Fig. 2). Only 16O/o of the fields were smaller than 100 deg2, and the smallest one was 3 deg2. The fields over 2,000 deg2 were also rare (gola) and the largest had 6,400 deg*. Fig. 2. The distribution of receptive field size in the 446 investigated cells. Ordinate, number of cells in the group; abscissa, receptive field area (the result of multiplication of its two axes). In the first group the subgroup of small fields is shown. A correlogram of the receptive field center deviation and field size was drawn (Fig. 3). The iilterpedence of the two measures appeared to be weak (Pearson's coefficient of linear correlation rxy = 0.099), but different from zero (Fischer's "t" test, p < 0.05 that r,, = 0). The correlation of the receptive field size and the displacement of its center from the vertical midline was also weak (rxy = 0.132). Thus there appears to be no general dependence of the receptive field size and eccentricity. But after classifying the areas of the fields as small (similar to the size of receptive fields in the visual cortex), medium and large (Table I) one can see that the distribution of deviations from center is different for the three groups. For these data, the X* = 42.2 (df = lo), which makes the probability of the homogenous distribution p < 0.0002. Thus near the visual center the small fields are more frequent, but because of their low number, they do not significantly influence the distribution. When the Clare'Bishop area had been mapped according to the procedure of each experiment, it appeared that in the center of the area units have contralateral receptive fields. The fields that extended across the midline to the ipsilateral side (30 /o of the fields) were found mostly at the periphery of the group. They were most frequent at the bottom of the sulcus but occurred also at the sides or above the contralateral

184 K. TURLEJSKI and A. MICHALSKI group. All the fields were very close to. the midline and only three of them were entirely ipsilateral. Ophthalmic examination showed that the left eye was sometimes displaced slightly leftward probably because of removing the masseter muscle. This could influence the position of the receptive fields in relation to the midline, but more than 70 /o of the fields passed across the midline farther than could be caused by such an artefact. In spite of checking all the contralateral group no strong representation of the upper field was found; 78O/o of receptive field centers were placed beneath deg 6000 - aaoo - 4000-3.000-2,000 -................... 1000 -..,...... :................:.:...::. : :......,...................,.^.............:.:.....: :.....,.... ':...,... : : :,... ;.... 100.. : A:... -..,.... :.......,,... 1 1 1 I 0 10 20 30 40 50 60 70 80 dey lw 1 Fig. 3. The correlogram of the receptive field size and its center deviation (w). Data for 446 cells. Every point represents one receptive field. Ordinate, area of the field, abscissa deviation of the field center.

CLARE-BISHOP AREA IN THE CAT Distribution of receptive field center deviations in the groups of small, medium and large fields Field size 1 Distance from center (deviation) 0 10.5 20.5 30.5 40.5 50.5 > - -.-. - - - - - I I I I I I Small fields (1-100 deg2) Mediurnfields(l00-1,WOdeg2) Large fields (> 1,000 deg2) N 13 the Horsley-Clarke horizontal zero plane. The distribution of position angles (Fig. 4) emphasizes that the upper part of the visual field is represented less frequently than the lower. However, the large receptive fields, while having their centers in the lower visual field, covered a considerable part of the upper field too. 26-19 - 3-35 20 64 55-28 23 118 '- 8 2 2 ' 7 1 44 26 45 253-21 24 23 / 122 73 52 70 / 446-98 - -.- - Fig. 4. Distribution of the position angle (q) values of the receptive field centers. The numbei of cells having a value of the angle is proportional to the distance from the center of the radial scale to the polygon border in the direction of the value (scale in the Figure). The great prevalence of the values in between 90 and 180 deg can be seen. In most cases one part of the area represented a particular part OC the visual field, so retinotopy was preserved to some degree. But the individual variability of the retinotopy was too great to allow a more precise scheme to be drawn. DISCUSSION It is not clear whether units in the suprasylvian gyrus vs. suprasylvian sulcus differ significantly in their visual response (3, 4, 8, 10, 19)

186 K. TURLEJSKI and A. MICHALSKI and thus whether a boundary can be designated between the two populations. The only exception is that part of the suprasylvian sulcus representing the contralateral field (12), in opposition to the bilateral representation in the other parts (3-5). Palmer (14) supposed the Clare-Bishop area to extend all along the medial suprasylvian sulcus. However, he recorded responses from the contralateral visual field only in a part of the sulcus, and the part corresponds well to the presently described area. The other parts of the sulcus represented the central visual field. Thus the area of contralateral projection is more or less agreed upon in the literature, but the boundary of the Clare-Bishop area is undecided. We used binocular stimulation during experiments, because we wanted to find all the visually responsive cells, but this produced a disparity of ipsi- and contralateral receptive fields of cells. The shift to left and right was symmetrical, and then it could not change the position of the receptive field center. However, it could influence the computation of the receptive field area, especially for the small fields. The large fields without a clear border, that prevail in the Clare-Bishop area were not influenced very much. We confirmed most of Hubel and Wiesel's (12) data on retinotopy, such as the representation of the lower, contralateral visual field with large, widely scattered receptive fields. But the centripetal shift of receptive fields between the top and the bottom of the sulcus was observed in less than half of the penetrations, and this tendency was also absent in one of Hubel and Wiesel's penetrations (12). Our finding that units with the receptive fields crossing midline to the ipsilateral side were found in every part of the sulcus, not only in the bottom, is probably related to the Heath and Jones (9) finding that all parts of the su.lcus obtain callosal fibers. We found no strict dependence of field size upon distance from the center of the visual field, as postulated by Hubel and Wiesel (12), but there are somewhat different topographical distributions of the small and large receptive fields in this area. The large, widely scattered receptive fields make for an imprecise retinotopy in the Clare-Bishop area as compared to visual cortex (10, 11) thus resembling the lateral posterior nucleus (7). Therefore this area is unlikely to be primarily concerned with the precise localization of objects in the visual field. The prevalence of the lower visual field remains unexplained. We thank to Mrs. J. Rokicka for technical help in the experiments. This investigation was supported by Project 09.4.1 of the Polish Academy of Sciences and by Foreign Research Agreement 05.275.2 of the U. S. Department of Health. Education and Welfare under PL 480.

CLARE-BISHOP AREA IN THE CAT REFERENCES 1. BISHOP, P. O., KOZAK, W. and VAKKUR, G. J. 1962. Some quantitative aspects of the cat's eye: axis and plane of reference, visual field co-ordinates and optics. J. Physiol. (Lond.) 163: 466-502. 2. CLARE, M. H. and BISHOP, G. H. 1954. Responses from an association area secondarily activated from optic cortex. J. Neurophysiol. 17: 271-277. 3. DOW, B. M. and DUBNER, R. 1969. Visual receptive fields and responses to movement in an association area of cat cerebral cortex. J. Neurophysiol. 32: 773-784. 4. DOW, B. M. and DUBNER, R. 1971. Single-unit responses to moving visual stimuli in middle suprasylvian gyrus of the cat. J. Neurophysiol. 34: 47-55. 5. DUBNER, R. and EROWN, F. J. 1968. Response of cells to restricted visual stimuli in an association area of cat cerebral cortex. Exp. Neurol. 20: 70-86. 6. GLICKSTEIN, M., KING, R. A., MILLER, J. and BERKLEY, M. 1967. Cortical projections from the dorsal lateral geniculate nucleus of cats. J. Comp. Neurol. 130: 55-76. 7. GODFRAIND, J. M., MEULDERS, M. and VERAART, C. 1969. Visual receptive fields of neurons in pulvinar, nucleus lateralis posterior and nucleus suprageniculatus thalami of the cat. Brain Res. 15: 552-555. 8. GRUSSER, 0. J., GRUSSER-CORNEHLS, U. and HAMASAKI, D. 1972. The responses of single neurons of the visual association cortex of cats to moving stimuli. Pflii,g. Arch. 335 (Suppl.): R87. 9. HEATH, C. J. and JONES, E. G. 1971. The anatomical organization of the suprasylvian gyrus of ihe cat. Ergebn. Anat. Entwick1.-Gesch. 45: 1-64. 10. HUBEL, D. H. and WIESEL, T. N. 1962. Receptive fields, binocular interaction and functional architecture in the cat's visual cortex. J. Physiol. (Lond.) 160: 106-154. 11. HUBEL, D. H. and WIESEL, T. N. 1965. Receptive fields and functional architecture in two nonstriate visual areas (18 and 19) of the cat. J. Neurophysiol. 28: 229-289. 12. HUBEL, D. H. and WIESEL, T. N. 1969. Visual area of the lateral suprasylvian gyrus (Clare-Bishop area) of the cat. J. Physiol. (Lond.) 202: 251-260. 13. MARSHALL, W. H., TALROT, S. A. and ADES, H. W. 1943. Cortical response of the anesthetized cat to gross photic and electrical afferent stimulation. J. Neurophysiol. 6: 1-15. 14. PALMER, L. A. 1973. Extent and retinotopic organization of the Clare-Bishop area of the cat. Anat. Rec. 175: 406. 15. SHOUMURA, K. 1972. Patterns of fiber degeneration in the lateral wall of the suprasylvian gyrus (Clare-Bishop area) following lesions in the visual cortex in cats. Brain Res. 43: 264-267. 16. SHOUMURA, K. and ITOH, K. 1972. Intercortical projections from the latera! wall of the suprasylvian gyrus, the Clare-Bishop area, of the cat. Brain Res. 39: 536539. 17. THOMPSON, R. F., JOHNSON, R. H. and HOOPES, J. J. 1963. Organization of auditory, somatic sensory, and visual projection to association fields of cerebral cortex in the cat. J. Neurophysiol. 26: 343-364.

188 K. TURLEJSKI and A. MICHALSKI 18. TURLEJSKI, K. 1975. Visual responses of neurons in the Clare-Bishop area of the cat. Acta Neurobiol. Exp. 35: 189-208. 19. WRIGHT, M. J. 1969. Visual receptive fields of cells in a cortical area remote from the striate cortex in the cat. Nature 223: 973-975. 20. ZERNICKI, B. 1968. Pretrigeminal cat. Brain Res. 9: 1-14. Received 22 October 1974 Krzysztof TURLEJSKI and Andrzej MICHALSKI, Department of Neurophysiology, Nencki Institute of Experimental Biology, Pasteura 3, 02-093 Warszawa, Poland.