Brain Research, !17 (1976) 369-386

© Elsevier/North-Holland Biomedical Press, Amsterdam - Printed in the Netherlands

369

Research Reports

F U R T H E R OBSERVATIONS ON CORTICOFRONTAL CONNECTIONS IN THE RHESUS MONKEY*

DOUGLAS A. CHAVIS and DEEPAK N. PANDYA Bullard attd Denny-Brown Laboratories Harvard Neurological Unit, Beth Israel Hospital Boston, Mass. 02215, The Aphasia Research Center, Deparlment of Neurology, Boston University Medical School, Boston, Mass., and The Bedjbrd Veterans Administration Hospital, Bed/brd, Mass., 02130 (U.S.A.)

(Accepted April 5th, 1976)

SUMMARY The frontal lobe connections of the post-Rolandic sensory association areas are investigated. Our results indicate that the caudal portion of the superior temporal gyrus (area 22), the lateral peristriate belt (areas 18, 19), and the superior parietal lobule and the rostralmost portion of the inferior parietal lobule (areas 5 and 7), all project to periarcuate cortex, while the middle portion of area 22, caudal inferotemporal cortex (area 20), and the middle portion of the lower bank of the intraparietal sulcus, all have connections predominantly to prearcuate cortex. In contrast, rostral area 22 and the rostral inferotemporal cortex (area 21 ) project primarily to the orbital surface, and the middle portion of area 7 projects to the mid-principal sulcus. Those regions ,,hat project to periarcuate cortex are termed first association areas (AAI, VAI, SAI), those that project primarily to prearcuate cortex are designated second association areas (AA2, VA2, SA2), while those that project mainly to the orbital surface or the mid-principal sulcus are called third association areas (AA3, VA3, SA3). It was also found that the caudalmost portion of area 7 has a distinct projection pattern, connecting with the dorsal prearcuate cortex .... areas 8B and 46. Additionally, it was observed that the connections from the association areas of different sensory modalities appear to overlap in specific areas of frontal cortex. Projections from the first association areas seem to converge in the periarcuate zone (bimodal overlap is noted between VAI and AA! in the arcuate concavity, and SA1 and VAI dorsal to the arcuate sulcus), while those from the second association areas * A preliminaryreport of this investigationhas been publishedelsewhere4.

370 overlap in the ventral prearcuate cortex (area 46), where both bimodal and trimodal overlap is observed.

INTRODUCTION Previous studies of cortico-cortical connections in the rhesus monkey have found a stepwise progression of projections originating in primary sensory cortex, extending outward into the association areas of each modaiity, and from these into more distal areas, such as the frontal lobetS,17,3~,a6. It has thus been established that the posterior parietal lobe, the lateral peristriate cortex, and the superior and inferior temporal gyri send fibers to the frontal 1obe~5,17,3~,a6. Since some of these fiber systems were observed to terminate in close proximity to each other, it was suggested that there might be 'polysensory' or sensory 'convergence' zones within frontal cortexlS,3~,43,46,. The location of these convergence centers, and the type of anatomical input they receive, have not, however, been established with certainty. Indeed, the precise origin of all the frontal lobe connections and their exact termination fields remain to be delineated. The present study, therefore, continuing previous work, was focused on the projections to the frontal lobe from the post-Rolandic association areas in order to establish the precise termination fields of these projections, and consequently has also clarified to some extent the location of, and inputs into, the regions of presumed sensory convergence. MATERIALS AND METHODS The 31 cases examined in this study were drawn from a series of rhesus monkey experiments already available in our laboratory, in each case a single cortical lesion had been produced by subpial aspiration in one of the various parts of the superior temporal, peristriate, inferotemporal, and parietal cortical association areas (Fig. 2). Seven to I 1 days following surgery, the animals had been deeply anesthetized and transcardially perfused with saline followed by 10% formalin, and the brains subsequently had been processed for staining by the Fink-Heimer and Nauta-Gygax methods 7,2~. The anterograde fiber degeneration was charted with an X-Y recorder, and the data reconstructed on equal size tracings of photographs of the brains. The depths and the banks of the principal and arcuate sulci were portrayed on these tracings in two dimensions in a uniform fashion. Areas of overlapping projections were ascertained by superimposing the tracings on one another, giving special attention to topographical landmarks such as the bottom and crown of sulci. To further clarify suggested areas of overlap, the results obtained by superimposition were then compared with the transverse sections of the charted degeneration, at each respective level. RESULTS In presenting the results of this study, the post-Rolandic association cortex will be divided into 3 areas, i.e., the superior temporal gyrus, peristriate-inferotem-

371

PERIARCUATE

A

I PREARCUATE

Fig. 1. Architectonic parcellations of cortical areas. In A, P~ 9dmann's numerical schema, Sanides' architectonic map of the superior temporal gyrus, and the ar ~:oximate boundaries of the periarcuate and prearcuate regions of the frontal lobe are shown, as referred to in the text. In B, Walker's maps of the frontal lobe are shown. poral cortex, and the posterior parietal lobules. Brodmann's numerical schema 3 will be used in referring to the general architectonic designation of the cerebral cortex (Fig. IA). In addition, Walker's architectonic classification 4T will be employed for the frontal lobe (Fig. I B), while the recently described parcellation of Sanides a2 will be used for the superior temporal gyrus (Fig. IA). With regard to nomenclature, the term 'periarcuate cortex' will refer to the cortical region in and around the arcuate sulcus (areas 6, 8, and 45). The term 'prearcuate cortex' will refer to the prefrontal cortex rostral to the arcuate sulcus on the lateral surface (Fig. I A), while 'orbital cortex' refers to the prefrontal cortex overlying the orbit. Although many experimental cases were studied (Fig. 2), only the frontal lobe connections of representative cases will be described. While most of our lesions were confined to the cerebral cortex, some extended into the underlying white matter. Our conclusions, however, are derived from a comparison of many different lesions, both with and without white matter involvement.

Afferentsfrom the superior temporalgyrus Seven ablations of various subdivisions of the superior temporal gyrus were examined (Fig. 2, cases 1-7). Chartinbs of the fiber degeneration revealed 3 major patterns of frontal lobe afferents, each characterized by the relative predominance of its distribution to the periarcuate, prearcuate, or orbital region, respectively. Following lesions of the caudalmost sector of the superior temporal gyrus, areas Tpt and Pa-Alt (Figs. IA and 3A, case 1), dense fiber degeneration in the frontal lobe was observed in the dorsal arcuate concavity caudal to the tip of the principal sulcus (area 8A). This dege~.-~ration extended into the adjacent bank of the arcuate sulcus, and, with lesser intensity, along its inferior limb. Additional terminal degener-

CS

:AS( LF

/

STS

®

® ® f - J

LF j

STS

Fig. 2. Composite diagram to show the extent of various lesions of the temporal lobe (superior and inferior temporal gyri), parietal lobe, and occipital lobe association areas. Note that the lesions in cases 17 and 23 are placed within the banks of the intraparietal sulcus.

Case 1

%::!¢ /kd.

A

B

C-ase 4

Case 6 .. :!.i}" i ~

/3

/

.,,. /

Fig. 3. Diagrammatic representation of three lesions of the superior temporal gyrus (diagram A is a composite diagram to show the extent of 3 lesions: case 1, involving the caudal portion, case 4 the middle portion, and case 6 the rostral portion of the gyrus) and resulting degeneration, on the lateral surface of the frontal lobe (shown by dots), from these cases (B, C and D, respectively). In C and D, the orbital surfaces are also illustrated to show the degeneration. (Note that in this and subsequent Figs. 4 and 5, the arcuate and principal sulci are opened in the insets to expose the depths.)

373 ation was noted in area 6 dorsal to the arcuate sulcus. Only sparse degeneration was seen in prearcuate cortex (Fig. 3B, case l ). Ablations of the central sector of the superior temporal gyrus, areas Pa-Alt and T~3, produced dense degeneration in the same periarcuate area (Figs. I A and 3A, case 4), but also elicited substantial terminal degeneration in prearcuate cortex, in particulal in the mid-rostral portion of the upper bank of the principal sulcus, the adjacent dorsolateral cortex (dorsal areas 46 and 8B), as well as in the region between the arcuate and fronto-orbital sulci (areas 45 and 46). This latter field of degeneration was more extensive than that found in the cases involving the caudal area. Only minor degeneration was observed in area 12 on the orbital surface in this case (Fig. 3C, case 4). Unlike more caudal and central ablations, lesions in the rostral portion of the superior temporal gyrus, area Ts2 (Figs. 1A and 3A, case 6), produced extensive terminal degeneration in areas 12 and 13 on the orbital surface of the frontal lobe. As in the case of Pa-Alt and Ts3 ablation, degenerated terminals were also seen in the rostral portion of the principal sulcus, but in this case of more rostral lesion it remained primarily within, and extended to the rostral tip of, the sulcus (Fig. 3D, case 6). In addition, a small quantity Gf degeneration appeared in the ventrolateral aspect of area 46, as well as in the region just rostral to the superior limb of the arcuate sulcus, and on the medial surface of the prefrontal cortex. It is evident from these findings that the caudal portion of the superior temporal gyrus, areas Tpt and Pa-Alt, projects to periarcuate cortex, whereas the rostral area, Ts2, projects most strongly to the orbital surface, and the middle area, Pa-Alt-Ts3, has projections characteristic of both but connects most heavily with the prearcuate cortex.

Afferents fi'om peristriate and #~ferotemporal cortex Seven lesions in the peristriate and inferotemporal association cortex, encompassing areas 18, 19, 20 and 21, were examined (Figs. IA and 2, cases 8-14). Analysis of these cases, likewise, revealed 3 distinct patterns of efferent association with the frontal lobe. Lesions of the preoccipital gyrus (areas 18 and 19) caused terminal degeneration in the rostral bank of the arcuate sulcus and adjacent ventral arcuate concavity. When the lesion involved area 18 and the caudal part of area 19 on the lateral surface (Fig. 4A, case 9), the degeneration was restricted more to the cortex lining the rostral wall or the arcuate sulcus (Fig. 4C, case 9). Those lesions which extended more rostrally in area 19 (Fig. 4A, case 11 ) produced fiber degeneration not only in the anterior wall of the arcuate sulcus, but also in the lower arcuate concavity extending forward to the caudal part of the principal sulcus, areas 8 and 45 (Fig. 4D, case I ! ). Lesions in the most dorsal portion of area 19 (Fig. 4A, case 8), however, sharply differed from the more ventrally placed lesions with respect to the resulting pattern of fiber degeneration in the frontal lobe. Lesions in this dorsalmost region produced degeneration in the rostral bank of the superior limb of the arcuate sulcus, extending into area 8B, and in the depth of the principal sulcus (Fig. 4B, case 8).

374 Unlike the periarcuate concentration of terminal degeneration elicited by lesions in peristriate regions, lesions in area 20 of the inferotemporal cortex (Fig. 4A, case 12) caused degeneration predominantly in the prearcuate region. Most of this degeneration was distributed in area 46 below the caudal portion of the principal sulcus and extended to the fronto-orbital sulcus (Fig. 4E, case 12). A small amount of terminal degeneration was also noted in the ventral arcuate concavity in this case. The pattern of degeneration resulting from lesions in the rostroventral portion Case 8

I" Case 9

Case 11

(/ ¢.5 j

,P.S./ A.S I

tL.F.

/

l~

/ ~.s,"

ST 5

/_ Case 12

Case 14

,.S.~..~•.":";':~~" --A.s.

/'~

'

cs

~

o'"

--Y'W

STS



'. •



f...~--~.' F'½

~

,j,,

/

Fig. 4. Diagrammatic representation of 5 lesions of the peristriate and inferotemporal areas (diagram A is a composite diagram to show the extent of these lesions; cases 8, 9 and ! 1 are in the peristriate belt while cases 12 and 14 are in inferotemporal cortex) as well as resulting degeneration from these cases (B, C, D, E and F, respectively)on the lateral surface of the frontal lobe. In F, the orbital surface is illustrated to show the degeneration.

375 of inferotemporal cortex, area 21 (Fig. 4A, case 14), sharply contrasted with that from the previous cases by producing heavy degeneration in the orbitofrontal cortex, primarily in area i I. Such lesions also produced fiber degeneration in area 46 below the middle portion of the principal sulcus, and in a small region of the cortex forming the anterior wall of the inferior limb of the arcuate sulcus (Fig. 4F, case 14). In conclusion, the lateral peristriate cortex projects to the periarcuate region*, while caudal inferotemporal cortex projects to the prearcuate area, and the rostral inferotemporal cortex projects to mainly the orbital surface of the frontal lobe.

Afferents fi'om the parietal association cortex An examination of 11 lesions involving parietal association cortex areas 5 and 7 (Figs. 1A and 2, cases 15-25) revealed 4 general patterns of parietofrontal association. Lesions of the superior parietal lobule and the rostralmost portion of the inferior parietal lobule (Fig. 5A, cases 15, 16, 17 and 18), resulted in fiber degeneration in area 6, the premotor cortex. Both rostral and caudal lesions of area 5 (Fig. 5A, cases 15 and 16) produced degeneration in the dorsolateral part of area 6, around the superior precentral dimple. These lesions differed from each other, however, in that the rostrai lesion (Fig. 5C, case 16) produced additional degeneration in MIl, the supplementary motor area on the medial surface, while the caudal lesion (Fig. 5B, case 15) elicited degeneration in the caudal bank of the arcuate sulcus, in case 16 some additional degeneration was found in area 8B; this, however, could be attributed to a small area of damage in the adjoining ventral bank of the intraparietal sulcus (see Fig. 5J, case 25). When the lesion involved the upper bank of the intraparietal sulcus, as in Fig. 5A, case 17, degeneration was again found in Mil and in the dorso!ateral part of area 6, but it additionally continued into the spur of the arcuate suicus and below it, into the ventral region of area 6 (Fig. 5D, case 17). When the rostralmost portion of the inferior parietal lobule was ablated (Fig. 5A, case 18), degeneration in the frontal lobe was found to be restricted to the ventrolateral region of area 6 and the adjacent caudal bank of the inferior limb of the arcuate sulcus, as illustrated in Fig. 5E, case 18. In contrast to the premotor distribution of degeneration observed to result from lesions of the rostralmost part of the inferior parietal Iobule, lesions of the remainder of the lobule (Fig. 5A, cases 19, 20, 21, 23 and 25) caused terminal degeneration in both periarcuate and prearcuate cortex. The largest of these lesions (Fig. 5A, case 19), which damaged most of the inferior parietal Iobule without encroaching upon the intraparietal sulcus, produced degeneration in the middle portion of the principal sulcus, dorsal areas 46 and 8B, the upper bank of the superior limb of the arcuate sulcus and the rostral bank of the lower limb (Fig. 5F, case 19). Some concomitant degeneration in the caudal bank of the lower limb of the arcuate sulcus presumably resulted from an extension of the lesion into the most rostral part of area 7, as in Fig. 5E, case 18. Analysis of the varying patterns of the degeneration resulting from more limited lesions in both the convexity of area 7 and the adjacent intraparietal * It should be pointed out that preliminary observations suggest that medial and ventral portions of peristriate cortex do not project to the periarcuate region.

..;:: \

/

) .,/'~ !

'S V

l i5 l l l l t ; I

U C ; wilwili

,.~','" ..s~I

:7[~"

Oi l u ~ ~;,,ll

I .'f I ~'--- ,.-,,~'/" "#~ ' ~

/

f

s~s//

'

.,.i/

dl /

3'1 ~"

J

.!

I

I I li l l l l l ; l l t

~-;:)

9 I llseO

k~ " ~ _4/; ---.~:~:,'l

Further observations on corticofrontal connections in the rhesus monkey.

Brain Research, !17 (1976) 369-386 © Elsevier/North-Holland Biomedical Press, Amsterdam - Printed in the Netherlands 369 Research Reports F U R T...
2MB Sizes 0 Downloads 0 Views