NeuroImage 19 (2003) 959 –967
www.elsevier.com/locate/ynimg
When vision guides movement: a functional imaging study of the monkey brain Georgia G. Gregorioua,b and Helen E. Savakia,b,* a
Department of Basic Sciences, Faculty of Medicine, School of Health Sciences, University of Crete, Crete, Greece Institute of Applied and Computational Mathematics, Foundation for Research and Technology, Hellas, Greece
b
Received 30 October 2002; revised 7 February 2003; accepted 19 February 2003
Abstract Goal-directed reaching requires a precise neural representation of the arm position and the target location. Parietal and frontal cortical areas rely on visual, somatosensory, and motor signals to guide the reaching arm to the desired position in space. To dissociate the regions processing these signals, we applied the quantitative [14C]-deoxyglucose method on monkeys reaching either in the light or in the dark. Nonvisual (somatosensory and memory-related) guidance of the arm, during reaching in the dark, induced activation of discrete regions in the parietal, premotor, and motor cortices. These included the dorsal part of the medial bank of the intraparietal sulcus, the ventral premotor area F4, the dorsal premotor area F2 below the superior precentral dimple, and the primary somatosensory and motor cortices. Additional parietal and premotor regions comprising the ventral intraparietal cortex, ventral premotor area F5, and the ventral part of dorsal premotor area F2 were activated by visual guidance of the arm during reaching in the light. This study provides evidence that different regions of the parieto–premotor circuit process the visual, somatosensory, and motor-memory-related signals which guide the moving arm. © 2003 Elsevier Science (USA). All rights reserved.
Introduction How does the brain produce a reaching movement which brings the arm accurately to an object? When the object is visible, reaching is achieved by utilization of (1) visual input, which provides information about the location of the target in the extrapersonal space and its relative distance from the arm, and (2) somatosensory (proprioceptive) input, which provides information about the position of the moving arm. Reaching in the dark is achieved by utilization of (1) somatosensory information about the position of the arm, and (2) sensory-motor memories of an internal representation of the target location and the motor act. In brief, visual and somatosensory inputs are available to guide the moving arm during reaching in the light toward visible objects, whereas somatosensory and memory-related infor-
* Corresponding author. Department of Basic Sciences, Faculty of Medicine, School of Health Sciences, University of Crete, P.O. Box 1393, GR-71110, Iraklion, Crete, Greece. Fax: ⫹30-2810-394530. E-mail address:
[email protected] (H.E. Savaki).
mation is available to guide the arm during reaching in the dark. The intraparietal and premotor cortices are known to use visual, somatosensory, and motor information to guide the moving arm (Mountcastle et al., 1975; Bioulac and Lamarre, 1979; Hyvarinen, 1982; Kalaska et al., 1983; Georgopoulos and Massey, 1985; Colby and Duhamel, 1991; Caminiti et al., 1996; Fogassi et al., 1996; Kalaska, 1996; Graziano et al., 1997; Snyder et al., 1997; Wise et al., 1997; Rizzolatti et al., 1998; Savaki and Dalezios, 1999). However, the precise parieto–frontal regions, which process the different types of sensory-motor inputs used to guide the movement, are not known. Moreover, it is unknown if and where these inputs are combined in order to provide an integrated sense of the moving arm. In the present study, we used the [14C]-deoxyglucose (14C-DG) quantitative autoradiographic method (Sokoloff et al., 1977) to map the parieto–frontal cortex of monkeys reaching either to visual targets in the light or to memorized targets in the dark. Our findings demonstrate that there are distinct regions in the parietal and premotor cortices which are selectively activated by either visual or somatosensory
1053-8119/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S1053-8119(03)00176-9
960
G.G. Gregoriou, H.E. Savaki / NeuroImage 19 (2003) 959 –967
and memory-related guidance of reaching. They also indicate that the visual and nonvisual signals used to guide the reaching arm remain mostly segregated throughout this circuit.
Materials and methods Six head-fixed adult female rhesus monkeys weighing between 3 and 4 kg were used in accordance with a protocol approved by the Greek Veterinary authorities and the FO.R.T.H. animal use committee, and complying with European Union directive 86/609. Behavioral tasks The behavioral apparatus was placed 23 cm in front of the monkeys. Visual targets were red spots of 1.5° diameter. Visually guided saccades and reaching movements (with the left forelimb) were made from a central position (in the median sagittal plane at shoulder height) to an up-left peripheral position (at a distance of 20° and in a direction of 45°). Monkeys working in the light were required to hold eye and finger position within circular windows of 2.5° and 1.5 cm diameter, respectively, around the visual targets. The monkey working in the dark was required to hold eye and finger position within circular windows of 5° and 3.5 cm diameter, respectively, around the memorized positions. Eye position was recorded with scleral search coils as previously described (Moschovakis et al., 2001). Monkeys performed their tasks for 45 min during the 14C-DG experiment and received water as reward. Two pairs of monkeys were used to study the effects induced by reaching in the light. The first pair included (1) an experimental monkey which performed reaching movements from a central to a peripheral (20° up-left) visual target while its gaze remained fixed on the central target (RLf, reaching in the light while fixating), and (2) its control monkey, which maintained its gaze fixed on the central visual target (Cf, fixation control). The RLf monkey had to fixate an illuminated central target and touch it with the index of its left forelimb for 0.8 –1.5 s, until the peripheral target was illuminated. Then the monkey had to reach and touch the peripheral target and hold it for 0.5–1 s, while maintaining its gaze fixed on the central visual target. Intertrial intervals were 1–1.8 s long. This monkey performed 50 reaching movements during the critical 5 first min of the 14 C-DG experiment. Its control, Cf monkey, which had to maintain its gaze fixed on the central visual target during the period it was illuminated (4 s per trial), maintained fixation for 75% of the experimental period. Intertrial intervals were 0.2– 0.3 s long. The second pair included (1) an experimental monkey which performed the same (20° up-left) visually guided reaching movements during natural unconstrained oculomotor behavior, i.e., while executing saccades (RLs, reaching in the light while saccading), and (2) its control
monkey, which executed visually guided saccades 20° upleft (Cs, saccades control). The RLs monkey performed 45 reaching movements during the critical 5 min. The difference from the task of the RLf monkey was that the central target was turned off when the peripheral target was illuminated. Its control, the Cs monkey, had to fixate the illuminated central target for 0.5–1 s until it was turned off. Then the peripheral target was illuminated, signaling a saccade within 1 s and fixation for 0.5–1 s until it disappeared. Intertrial intervals were 1–1.8 s long. This monkey performed 60 saccades during the critical 5 min. A third pair of monkeys was used to study the effects induced by reaching in the dark. The experimental monkey performed acoustically triggered arm reaching movements from a memorized central to a memorized peripheral location (20° up-left) in complete darkness while its eyes maintained a straight ahead direction (RD, reaching in the dark), and its control monkey remained in the dark (Cd, dark control). The RD monkey, following an auditory cue (90 Hz), had to look straight ahead toward a memorized location corresponding to the central position, to reach (within 3 s) and touch the screen at this central position and hold it for 0.6 –1 s. Then a second auditory cue (180 Hz) signaled a reaching movement (within 2 s) to the memorized peripheral position and holding of this position (for 0.5–1 s), while the eyes maintained the straight ahead direction. Intertrial intervals were 0.5– 0.9 s long. View was completely occluded by a black barrier in front of the monkey’s eyes. This monkey performed 40 reaching movements during the critical 5 min. The Cd monkey was an untrained control, seated in front of the nonfunctioning behavioral apparatus in the dark, and receiving neither sensory stimuli nor liquid reward during the 14C-DG experiment. [14C]Deoxyglucose maps and statistics The 14C-DG experiment and the brain tissue processing for autoradiography were performed as previously described (Savaki et al., 1993; Gregoriou and Savaki, 2001). Two-dimensional (2D) reconstruction of the spatio-intensive pattern of metabolic activity (LCGU values in mol/ 100 g/min) within the rostrocaudal and the dorsoventral extent of each cortical area of interest was generated in each hemisphere from horizontal sections as previously described (Dalezios et al., 1996). The total of 550, 640, 610, and 990 serial sections of 20-m thickness were used in each hemisphere for the intraparietal, the ventral premotor, the dorsal premotor, and the primary somatosensory-motor cortex, respectively. A precise anatomical point was used for the alignment of adjacent sections, e.g., the intersection of the intraparietal with the parietoccipital sulcus for the map of the intraparietal cortex, and the anterior crown of the central sulcus for the map of the dorsal premotor cortex. The plotting resolution of the maps was set to 100 m. The average of glucose utilization values was calculated in sets of five adjacent sections throughout each cortical
G.G. Gregoriou, H.E. Savaki / NeuroImage 19 (2003) 959 –967
961
Table 1 Metabolic effects in subregions of the intraparietal, premotor, primary motor, and somatosensory cortices Cortical area
Intraparietal Dorsal area 5 Ventral intraparietal Premotor F2-dimple F2-periarcuate F4a F5-bank F5-convexity Somatosensory-motor SI-forelimb F1-forelimb
RLf
RLs % Ipsi
RD
% Ipsi
% Contra
% Contra
% Ipsi
% Contra
⫺2 9
19 23
0 7
25 13
5 7
31 4
2 13 9 8 20
17 27 11 22 37
4 14 7 17 10
22 17 11 18 18
7 0 16 10 4
24 4 20 12 4
17 6
51 43
12 9
46 52
33 8
70 45
Note. RLf, monkey reaching in the light while fixating; RLs, monkey reaching in the light while saccading; RD, monkey reaching in the dark. % Ipsi, ipsi-to-control percentage difference calculated as [(Ipsi-Left)/Left] ⫻ 100, where Ipsi is the hemisphere ipsilateral to the moving left forelimb, and Left is the left hemisphere of the control monkey. % Contra, contra-to-control percentage difference calculated as [(Contra-Right)/Right] ⫻ 100, where Contra is the hemisphere contralateral to the moving left forelimb, and Right is the right hemisphere of the control monkey. All differences above 10% were statistically significant by both the Student’s unpaired t test and the Kolmogorov–Smirnov test at the level of P ⬍ 0.01. a The ventralmost part of F4 which corresponds to the mouth representation (Gentilucci et al., 1988) has been excluded.
area. Normalization of LCGU values was based on the average unaffected gray matter value pooled across all monkeys. Experimental-to-control and contra-to-ipsi glucose utilization values were compared for statistical significances by the Student’s unpaired t test. Given that LCGU values for homologous areas in the two hemispheres of a normal resting monkey differ by up to 7% (Savaki et al., 1993) we adopted a conservative criterion accepting as significant only differences exceeding 10% and reaching the 0.01 level of confidence. The Kolmogorov–Smirnov test was also used to check for significant differences in the distribution of values. The individual intraparietal 2D maps of activity in 6 of the 12 hemispheres presently analyzed have been previously
presented (Gregoriou and Savaki, 2001). Here, control-toexperimental and contra-to-ipsi statistical comparisons (Table 1) as well as critical subtractions among the 12 hemispheres (Figs. 1– 4) reveal the precise (1) intraparietal, (2) dorsal premotor, (3) ventral premotor, and (4) primary motor cortical regions specifically associated with the visual and the nonvisual guidance of movement. Geometrical normalization Geometrical normalization of the intraparietal 2D maps was based on surface landmarks (the fundus and the crowns of the sulcus). One of the intraparietal maps was considered as reference. The remaining maps were manipulated with
Fig. 1. Activations in the intraparietal cortex induced by reaching in the light and reaching in the dark. The white dashed line represents the fundus of the intraparietal sulcus whereas the two dotted lines represent the two crowns. Caudal limit delineates the intersection of the intraparietal with the parietoccipital sulcus. Color bars indicate the difference in local cerebral glucose utilization (LCGU, in mol/100 g/min). (A) The experimental-to-control significant differences induced by reaching in the light (RLf-contra minus Cf) were located in the ventral intraparietal areas 5 and 7 (around the fundus) and in the dorsal intraparietal area 5. (B) The map of contra-to-ipsi differences (RLf-contra minus RLf-ipsi) fails to illustrate only the activations around the caudal part of the fundus (compare with A) which are poorly lateralized. (C) The experimental-to-control significant differences induced by reaching in the dark (RD-contra minus Cd) were located in the dorsal part of the intraparietal area 5. (D) The map of contra-to-ipsi differences (RD-contra minus RD-ipsi) is almost identical to that of the experimental-to-control differences (C) due to the lateralization of effects. (E) Image generated after subtraction of C from A, in order to reveal the intraparietal regions involved exclusively in visual guidance of reaching. (F) The image generated after the subtraction RLf-contra minus RD-contra is very similar to that of E. (G) Side view of the right hemisphere of a monkey brain illustrating the unfolded intraparietal cortex (outlined in red) shown in the reconstructed metabolic maps. C, caudal; D, dorsal; R, rostral. Fig. 2. Activations in the dorsal premotor cortex induced by reaching in the light and reaching in the dark. The thick solid white line rostrally represents the anterior tip of the brain in dorsal sections in which the arcuate sulcus is absent, and the anterior crown of the arcuate sulcus in ventral sections. Caudal limit delineates the anterior crown of the central sulcus. White thin solid lines represent cytoarchitectonic borders of area F2. Upper dotted line represents the superior precentral dimple. Three lower dotted lines from rostral to caudal: anterior and posterior tip of the floor and posterior crown of the arcuate sulcus. (A) The significant premotor activations induced by reaching in the light (RLf-contra minus Cf) were located in the ventral (periarcuate) and dorsal (around the superior precentral dimple) parts of area F2. (B) The single significant premotor activation induced by reaching in the dark (RD-contra minus Cd) was located in the dorsal part of area F2. (C) Image generated after subtraction of B from A, in order to reveal the dorsal premotor regions involved exclusively in visual guidance of reaching. (D) Side view of the right hemisphere of a monkey brain illustrating the reconstructed dorsal premotor areas outlined in red. The arcuate sulcus is unfolded to unveil the anterior and posterior banks of the superior limb as reconstructed in the maps. V, ventral. Other conventions as in Fig. 1.
962
G.G. Gregoriou, H.E. Savaki / NeuroImage 19 (2003) 959 –967
G.G. Gregoriou, H.E. Savaki / NeuroImage 19 (2003) 959 –967
963
the basis of known characteristics of areas SI, F1, F2, F4, F5, and F7 (Matelli et al., 1991; Geyer et al., 2000). For the dorsal premotor maps, the anterior crown of the central sulcus, the superior precentral dimple, and the cytoarchitectonic borders of F2 were used as landmarks. For the geometrical normalization of the ventral premotor maps we
Fig. 3. Activations in the ventral premotor cortex induced by reaching in the light and reaching in the dark. White solid lines represent cytoarchitectonic borders of areas F4 and F5. Dashed white line represents the posterior crown of the arcuate sulcus. The posterior bank of the arcuate sulcus and the exposed precentral cortex on the convexity lie on the right and the left side of the dashed line, respectively. (A) The main significant activations induced by reaching in the light (RLf-contra minus Cf) were located in area F5, both in the posterior bank of the arcuate sulcus and on the convexity. (B) The main significant activation induced by reaching in the dark (RD-contra minus Cd) was located within area F4. (C) Image generated after subtraction of B from A to reveal the ventral premotor regions involved exclusively in visual guidance of reaching. Alignment concerns areas F4 and F5. Regions outside their borders, such as SI, are not perfectly aligned. (D) Side view of the right hemisphere of a monkey brain illustrating the reconstructed ventral premotor cortex outlined in red. Arcuate sulcus is unfolded to demonstrate the cortex of the posterior bank reconstructed in the maps. Conventions as in Fig. 1.
linear transformations of the plane until we obtained the best fit among all the surface landmarks. The geometry of the arcuate sulcus was idiosyncratic due to the presence or lack of the floor and spur in the different hemispheres. Thus, geometrical normalization of the dorsal and ventral premotor cortical maps was based on cytoarchitectonic, in addition to the surface, landmarks. The different cytoarchitectonic subdivisions of the agranular frontal cortex were identified on Nissl-stained horizontal sections, on
Fig. 4. Activations in the cortex of the central sulcus induced by reaching in the light and reaching in the dark. White solid line represents the cytoarchitectonic border between the primary motor (F1) and primary somatosensory (SI) cortices. White dashed line represents the fundus of the central sulcus. (A) The most pronounced, statistically significant activations induced by reaching in the light (RLf-contra minus RLf-ipsi) were lateralized in the arm representation of both areas F1 and SI in the hemisphere contralateral to the moving arm. (B) The most pronounced, statistically significant activations induced by reaching in the dark (RDcontra minus RD-ipsi) were also lateralized in the arm representation of areas F1 and SI. (C) Activations due exclusively to visual guidance of reaching are not detected in the image generated after subtraction of maps RLf-contra minus RD-contra. (D) Side view of the right hemisphere of a monkey brain illustrating the reconstructed cortex outlined in red. Central sulcus is unfolded to unveil the cortex in its anterior and posterior banks. Conventions as in Fig. 1.
964
G.G. Gregoriou, H.E. Savaki / NeuroImage 19 (2003) 959 –967
transformed a cortical region extending from the fundus of the arcuate sulcus to the anterior crown of the central sulcus and from the dorsalmost border of area F4 to the ventralmost border of area F5. The entire dorsoventral extent of this region in each hemisphere was used as a reference to express the dorsoventral location of each section in the transformed map. The anteroposterior value of each pixel located either in the posterior bank of the arcuate sulcus or on the convexity between the arcuate and the central sulcus was calculated similarly. Consequently, rectangular isodimensional maps were generated in all hemispheres. The same procedure was followed for the geometrical normalization of the unfolded cortex of the central sulcus, in which case the transformed region extended from the anterior to the posterior crown of the sulcus. Emphasis was placed on the alignment of the relevant regions in each map. As a result, maps geometrically normalized for the dorsal and ventral premotor areas are not simultaneously normalized for adjacent cortical areas. Thus, Fig. 4 (normalized for the central sulcus) rather than Fig. 3 (normalized for ventral premotor areas) should be consulted for estimation of F1 and SI effects.
Results Intraparietal cortical areas The intraparietal regions activated during reaching in the light were elucidated after subtraction of the intraparietal map of the Cf monkey from the corresponding map of the RLf monkey (hemisphere contralateral to the moving arm) (Fig. 1A). The statistically significant activations (ventral ones around the fundus, and dorsal ones close to the crown of the medial bank) portray the combination of visual with nonvisual (somatosensory and motor-related) effects induced by reaching in the light. The dorsal activations were well lateralized in the hemisphere contralateral to the moving forelimb, in contrast to the ventral ones, which were poorly lateralized (Table 1). This quantitative finding is depicted in the image illustrating contra-to-ipsi differences (Fig. 1B, RLf-contra minus RLf-ipsi). Indeed, Fig. 1B resembles Fig. 1A (illustrating experimental-to-control differences) in the dorsorostral regions which display mostly lateralized effects, whereas it fails to demonstrate major activations in the ventral intraparietal cortex caudally around the fundus, i.e., in regions affected bilaterally (Gregoriou and Savaki, 2001). Apparently, between the intrasubject side-to-side and the intersubject experimental-to-control comparisons, the latter is more reliable when the effects are not completely lateralized. In any case, the close resemblance of Figs. 1A and B demonstrates how robust and reproducible the imaging data are when based on hundreds of serial sections analyzed quantitatively for glucose metabolism, and when high signal-to-noise ratio can be achieved in the individual maps (Gregoriou and Savaki, 2001).
The same regions were significantly affected in the RLs monkey (Table 1), and a similar image was obtained when we subtracted the intraparietal map of the Cs monkey from the map of the RLs monkey (hemisphere contralateral to the moving arm). The similar images of activations induced by reaching during experimentally constrained fixation (reaching while fixating) and during natural unconstrained oculomotor behavior (reaching while saccading) indicate that neither the angle of gaze nor the presence of saccades affect the spatial distribution of intraparietal effects elicited by visually guided reaching. Reaching in the dark induced a pronounced activation in the dorsal intraparietal area 5 (Table 1). The regions associated with nonvisual guidance of reaching were unveiled after subtraction of the map of the Cd monkey from that of the RD monkey (hemisphere contralateral to the moving arm). Two regions in the dorsal intraparietal area 5, close to the crown of the medial bank, were significantly activated (Fig. 1C). Intrasubject, contra-to-ipsi comparison (Fig. 1D), and intersubject, experimental-to-control comparison (Fig. 1C), provided very similar images. This similitude reflects the fact that the somatosensory and motor-related effects, elicited by nonvisually guided reaching, are significantly lateralized in the intraparietal cortex of the hemisphere contralateral to the moving arm (Table 1). Both visual and nonvisual information is used to guide reaching in the light, whereas only nonvisual information is available to guide reaching in the dark. Accordingly, to reveal the regions associated only with visual guidance of reaching, we subtracted the map illustrating the effects induced by reaching in the dark (Fig. 1C) from that illustrating the effects induced by reaching in the light (Fig. 1A). This generated image (Fig. 1E) is similar to that obtained after subtraction of RD-contra from RLf-contra (Fig. 1F), a fact which demonstrates once again the validity and power of our methods. These results indicate that the ventral intraparietal cortex is mainly implicated in the visual guidance of reaching (Figs. 1E and F) in contrast to the dorsal part of the intraparietal area 5 which is mainly involved in the nonvisual (somatosensory and motor-memory-related) guidance of the moving arm (Figs. 1C and D). It should be noted, however, that this functional segregation pertains to activation differences greater than 10%. Premotor cortical areas The primate premotor cortex has been anatomically subdivided into a dorsal and a ventral sector (Matelli et al., 1985; Barbas and Pandya, 1987). In our study, the effects in both these sectors were mostly bilateral, although the hemisphere contralateral to the moving arm was more affected than the ipsilateral one (Table 1). Consequently, intersubject comparisons of the activation maps are more reliable than the intrasubject ones. In the dorsal premotor cortex, reaching in the light while fixating (RLf-contra minus Cf) induced two significantly
G.G. Gregoriou, H.E. Savaki / NeuroImage 19 (2003) 959 –967
activated regions within the cytoarchitectonically identified area F2 (Matelli et al., 1991) (Fig. 2A). The first region is located in the dorsal part of F2 below the superior precentral dimple (F2-dimple) whereas the second one is ventrally located in the periarcuate area (F2-periarcuate). The same dorsal premotor regions were activated by reaching in the light while saccading (RLs-contra minus Cs). In contrast, only the F2-dimple region was significantly activated by reaching in the dark (Fig. 2B, RD-contra minus Cd). Given that there is no visual input in the latter case, the F2-dimple activation can only be attributed to somatosensory and/or motor-memory-related activity used to guide reaching in the dark. On the other hand, the F2-periarcuate activation which is significant in the RLf and RLs monkeys and absent from the RD monkey can only be attributed to visual information used to guide the forelimb in the light. Indeed, subtraction of the map in Fig. 2B (effects induced by reaching in the dark) from that in Fig. 2A (effects elicited by reaching in the light) revealed that only the F2-periarcuate region is associated with visual guidance of reaching (Fig. 2C). In the ventral premotor cortex, reaching in the light while fixating (Fig. 3A) and reaching in the light while saccading induced significant activations within areas F4 and F5 (Matelli et al., 1985; Geyer et al., 2000), including both F5-bank and F5-convexity (Table 1). Reaching in the dark induced significant activations in areas F4 and F5-bank, which can only be attributed to nonvisual information used to guide the reaching arm in complete darkness (Fig. 3B). Moreover, subtraction of Fig. 3B from 3A demonstrated that area F5 is mainly implicated in the visual guidance of forelimb movement (Fig. 3C), in contrast to area F4 which is mainly involved in the nonvisual (somatosensory and motor-memory-related) guidance of movement (Fig. 3B). The similar images of effects induced by reaching in the light while fixating and while saccading indicate that the spatial distribution of premotor activations elicited by visually guided reaching is not gaze-related. On the other hand, within the premotor regions associated mainly with visual guidance of movement (Table 1, F2-periarcuate, F5-convexity), the higher intensity of activations induced by reaching while fixating as compared to reaching while saccading may reflect the difference in gaze requirements. Indeed, strong gaze-related modulation of visual and reach-related discharge in the premotor cortex was reported during experimentally constrained fixation (Boussaoud et al., 1993; Mushiake et al., 1997; Boussaoud et al., 1998), whereas weak modulation was reported during natural unconstrained oculomotor behavior (Cisek and Kalaska, 2002). Cortical areas in the central sulcus Contra-to-ipsi intrasubject comparisons were used for the reconstructed cortical maps of the central sulcus because the forelimb effects in both the primary motor (F1) and somatosensory (SI) cortices were well lateralized (Table 1). Reaching in the light while fixating (RLf-contra minus RLf-
965
ipsi) induced significant activation around the middle of the dorsoventral extent of the sulcus (Fig. 4A). This activation extends into both the anterior and posterior banks (symmetrical around the fundus) and corresponds to the forelimb representation in F1 and SI cortices, respectively (Woolsey, 1958; Dalezios et al., 1996). Images similar to Fig. 4A were obtained from the monkey reaching in the light while saccading and from the monkey reaching in the dark (Fig. 4B), indicating that in F1 there is no segregation of visual and nonvisual (proprioceptive and memory-related) signals used for the guidance of movement. Indeed, Fig. 4C, illustrating the effects induced exclusively by visual guidance of reaching (RLf-contra minus RD-contra), does not reveal any significantly affected region in F1. The fact that the activation induced in the SI forelimb representation by reaching in the dark (Fig. 4B) is more intense and extended than that induced by reaching in the light (Fig. 4A, Table 1) indicates enhanced somatosensory processing when visual information is not available to guide the reaching forelimb. This is in accordance with earlier findings in visually deafferented monkeys, which have demonstrated that the SI cortex contralateral to a reaching forelimb becomes progressively more active as the visual input to that hemisphere is progressively diminished (Savaki et al., 1993). These findings indicate that the enhanced somatosensory activation in the monkey reaching in the dark acts as a compensatory mechanism for the absence of visual input.
Discussion This study provides quantitative pictorial representations of the intraparietal, premotor, and primary somatosensorymotor cortical regions processing either visual or nonvisual (somatosensory and motor-related) signals to guide the reaching arm. Our functional images demonstrate that reaching in the light and reaching in the dark engage distinct, circumscribed regions in the intraparietal cortex. Our predominant finding is that the neuronal populations related to the nonvisual (including somatosensory) and to the visual guidance of reaching remain mostly segregated in the dorsal and in the ventral part of the intraparietal cortex, respectively. These results are consistent with earlier anatomical studies according to which the ventral intraparietal cortex receives visual input (Maunsell and Van Essen, 1983; Ungerleider and Desimone, 1986), whereas more dorsal regions within the medial bank of the intraparietal sulcus receive input from somatosensory-related areas (Pandya and Seltzer, 1982). Moreover, the distribution of functional properties in the intraparietal cortex as revealed in our study is complemented by electrophysiological findings (Sakata et al., 1973; Mountcastle et al., 1975; Colby and Duhamel, 1991; Colby et al., 1993), and confirms previous metabolic and lesion reports (Savaki et al., 1993; Rushworth et al., 1997).
966
G.G. Gregoriou, H.E. Savaki / NeuroImage 19 (2003) 959 –967
However, in addition to providing evidence in favor of a functional distinction within the intraparietal sulcus, we here delineate in high-resolution maps the precise regions that are specifically engaged in the processing of visual and nonvisual signals during reaching. The existing, though sparse, overlap of intraparietal regions related to visual and nonvisual guidance of reaching may reflect the convergence of visual and somatosensory signals about static arm position seen in area 5 (Graziano et al., 2000). Downstream from the intraparietal cortex the next level implicated in the sensorial guidance of reaching is the premotor cortex (Godschalk et al., 1985; Wise, 1985; Caminiti et al., 1996; Fogassi et al., 1996, 1999, 2001; Wise et al., 1997; Matelli et al., 1998; Rizzolatti et al., 1998; Savaki and Dalezios, 1999). Our data demonstrate the existence of well-defined activations within both the dorsal and the ventral premotor cortices. In the dorsal premotor cortex, the functional dissociation of F2 into a dorsal region involved in nonvisually guided reaching and a ventral region implicated in visually guided reaching is supported by neurophysiological data demonstrating the presence of visually responsive neurons in the ventral part of F2 (Fogassi et al., 1999). Moreover, our finding that in both the intraparietal and dorsal premotor cortices the neuronal ensembles selectively involved in the nonvisual (including somatosensory) and visual guidance of reaching are segregated in their dorsal and ventral parts, respectively, is supported by anatomical data demonstrating that the dorsal F2 receives projections from the dorsal intraparietal area 5 which processes somatosensory information, whereas the ventral F2 receives projections from a more ventrally located intraparietal area (in the posterior half of the medial bank) which conveys mainly visual information (Matelli et al., 1998). The ventral premotor cortex also displayed circumscribed activations during reaching in light and in darkness. The sparse overlap of regions related to visual and nonvisual guidance of reaching may reflect the convergence of sensory signals about static arm position observed in the ventral premotor cortex (Graziano, 1999). On the other hand, our main finding that the neuronal populations related to nonvisual (including somatosensory) and visual guidance of movement remain mostly segregated in areas F4 and F5, respectively, is complemented by electrophysiological studies demonstrating preponderance of somatosensory-motor properties of neurons in area F4 (Gentilucci et al., 1988; Fogassi et al., 1996) and visuomotor properties of neurons in area F5 (Murata et al., 1997; Fogassi et al., 2001). Downstream from the premotor cortical centers, the next level of motor control, the primary motor cortex, was also activated by reaching. The similar effects induced in F1 by reaching in the light and reaching in the dark are compatible with previous data demonstrating that the primary motor cortex maintained a high activity level during both the visually instructed and the memorized delay periods of an arm movement (Smyrnis et al., 1992). Our functional images demonstrate that there is no segregation of visual and
nonvisual signals processed in F1 for arm guidance. The activated neuronal population in F1 may be associated with the motor act per se and/or the nonvisual (proprioceptive or memory-related) guidance of movement. All in all, our data demonstrate that discrete regions are activated in the intraparietal and premotor cortices during reaching in the dark. These regions are associated with the processing of proprioceptive-somatosensory and motormemory-related signals required to guide the moving arm in the dark. The activated premotor regions may project to F1 a program of the desired reaching movement, based on proprioceptive information about the limb position, and on motor memories related to the estimated distance between the felt position of the forelimb and the memorized internal representation of the target. Additional neuronal ensembles are engaged in both the intraparietal and premotor cortices during reaching in the light. The additional premotor neuronal ensembles may project to F1 signals, related to visuomotor transformations based on the distance between the position of the visible arm and the location of the visual target. These signals may contribute to a finer tuning of the population vector in F1 (Georgopoulos et al., 1986), thus enabling the generation of a more appropriate motor command from the upper motoneurons to secure a more accurate reaching movement. In conclusion, our data demonstrate that there are two largely segregated neuronal ensembles which selectively subserve visual and nonvisual guidance of reaching within the parieto–premotor pathway. This finding supports segregationist rather than integrationist theories of space perception for motor control.
Acknowledgments We are indebted to Massimo Matelli and Giuseppe Luppino for help in the identification of cytoarchitectonic borders. We also thank them, Apostolos Georgopoulos, and Adonis Moschovakis for constructive comments on an earlier version of the manuscript. The work was supported by Human Frontier Science Program (Grant RG0039/1988-B). G. Gregoriou was supported by the Greek General Secretariat of Research and Technology (Grant PENED/ 95ED24).
References Barbas, H., Pandya, D.N., 1987. Architecture and frontal cortical connections of the premotor cortex (area 6) in the rhesus monkey. J. Comp. Neurol. 256, 211–228. Bioulac, B., Lamarre, Y., 1979. Activity of postcentral cortical neurons of the monkey during conditioned movements of a deafferented limb. Brain Res. 172, 427– 437. Boussaoud, D., Barth, T.M., Wise, S.P., 1993. Effects of gaze on apparent visual responses of frontal cortex neurons. Exp. Brain Res. 93, 423– 434.
G.G. Gregoriou, H.E. Savaki / NeuroImage 19 (2003) 959 –967 Boussaoud, D., Jouffrais, C., Bremmer, F., 1998. Eye position effects on the neuronal activity of dorsal premotor cortex in the macaque monkey. J. Neurophysiol. 80, 1132–1150. Caminiti, R., Ferraina, S., Johnson, P.B., 1996. The sources of visual information to the primate frontal lobe: a novel role for the superior parietal lobule. Cereb. Cortex 6, 319 –328. Cisek, P., Kalaska, J.F., 2002. Modest gaze-related discharge modulation in monkey dorsal premotor cortex during a reaching task performed with free fixation. J. Neurophysiol. 88, 1064 –1072. Colby, C.L., Duhamel, J.R., 1991. Heterogeneity of extrastriate visual areas and multiple parietal areas in the macaque monkey. Neuropsychologia 23, 517–537. Colby, C.L., Duhamel, J.-R., Goldberg, M.E., 1993. Ventral intraparietal area of the macaque: anatomic location and visual response properties. J. Neurophysiol. 69 (3), 902–914. Dalezios, Y., Raos, V.C., Savaki, H.E., 1996. Metabolic activity pattern in the motor and somatosensory cortex of monkeys performing a visually guided reaching task with one forelimb. Neuroscience 72 (2), 325–333. Fogassi, L., Gallese, V., Buccino, G., Craighero, L., Fadiga, L., Rizzolatti, G., 2001. Cortical mechanism for the visual guidance of hand grasping movements in the monkey. A reversible inactivation study. Brain 124, 571–586. Fogassi, L., Gallese, V., Fadiga, E., Luppino, G., Matelli, M., Rizzolatti, G., 1996. Coding of peripersonal space in inferior premotor cortex (area F4). J. Neurophysiol. 76, 141–157. Fogassi, L., Raos, V., Franchi, G., Gallese, V., Luppino, G., Matelli, M., 1999. Visual responses in the dorsal premotor area F2 of the macaque monkey. Exp. Brain Res. 128, 194 –199. Gentilucci, M., Fogassi, L., Luppino, G., Matelli, M., Camarda, R., Rizzolatti, G., 1988. Functional organization of inferior area 6 in the macaque monkey. I. Somatotopy and the control of proximal movements. Exp. Brain Res. 71, 475– 490. Georgopoulos, A.P., Massey, J.T., 1985. Static versus dynamic effects in motor cortex and area 5: comparison during movement time. Behav. Brain Res. 18, 159 –166. Georgopoulos, A.P., Schwartz, A.B., Kettner, R.E., 1986. Neuronal population coding of movement direction. Science 233, 1416 –1419. Geyer, S., Matelli, M., Luppino, G., Zilles, K., 2000. Functional neuroanatomy of the primate isocortical motor system. Anat. Embryol. 202, 443– 474. Godschalk, M., Lemon, R.N., Kuypers, H.G.J.M., Van Der Steen, J., 1985. The involvement of monkey premotor cortex neurons in preparation of visually cued arm movements. Behav. Brain Res. 18, 143–157. Graziano, M.S.A., 1999. Where is my arm? The relative role of vision and proprioception in the neuronal representation of limb position. Proc. Natl. Acad. Sci. USA 96, 10418 –10421. Graziano, M.S.A., Cooke, D.F., Taylor, C.S.R., 2000. Coding the location of the arm by sight. Science 290, 1782–1786. Graziano, M.S.A., Hu, X.T., Gross, C.G., 1997. Visuospatial properties of ventral premotor cortex. J. Neurophysiol. 77, 2268 –2292. Gregoriou, G.G., Savaki, H.E., 2001. The intraparietal cortex: subregions involved in fixation, saccades, and in the visual and somatosensory guidance of reaching. J. Cereb. Blood Flow Metab. 21, 671– 682. Hyvarinen, J., 1982. Posterior parietal lobe of the primate brain. Physiol. Rev. 62 (3), 1060 –1129. Kalaska, J.F., 1996. Parietal cortex area 5 and visuomotor behavior. Can. J. Physiol. Pharmacol. 74, 483– 498. Kalaska, J.F., Caminiti, R., Georgopoulos, A.P., 1983. Cortical mechanisms related to the direction of two-dimensional arm movements: relations in parietal area 5 and comparison with motor cortex. Exp. Brain Res. 51, 247–260. Matelli, M., Govoni, P., Galletti, C., Kutz, D.F., Luppino, G., 1998. Superior area 6 afferents from the superior parietal lobule in the macaque monkey. J. Comp. Neurol. 402, 327–352.
967
Matelli, M., Luppino, G., Rizzolatti, G., 1985. Patterns of cytochrome oxidase activity in the frontal agranular cortex of the macaque monkey. Behav. Brain Res. 18, 125–136. Matelli, M., Luppino, G., Rizzolatti, G., 1991. Architecture of superior and mesial area 6 and the adjacent cingulate cortex in the macaque monkey. J. Comp. Neurol. 311, 445– 462. Maunsell, J.H.R., Van Essen, D.C., 1983. The connections of the middle temporal visual area (MT) and their relationship to a cortical hierarchy in the macaque monkey. J. Neurosci. 3, 2563–2586. Moschovakis, A.K., Gregoriou, G.G., Savaki, H.E., 2001. Functional imaging of the primate superior colliculus during saccades to visual targets. Nature Neurosci. 4, 1026 –1031. Mountcastle, V.B., Lynch, J.C., Georgopoulos, A.P., Sakata, H., Acuna, C., 1975. Posterior parietal association cortex of the monkey: command function for operations within extrapersonal space. J. Neurophysiol. 38, 871–908. Murata, A., Fadiga, L., Fogassi, L., Gallese, V., Raos, V., Rizzolatti, G., 1997. Object representation in the ventral premotor cortex (area F5) of the monkey. J. Neurophysiol. 78, 2226 –2230. Mushiake, H., Tanatsugu, Y., Tanji, J., 1997. Neuronal activity in the ventral part of premotor cortex during target-reach movement is modulated by direction of gaze. J. Neurophysiol. 78, 567–571. Pandya, D.N., Seltzer, B., 1982. Intrinsic connections and architectonics of posterior parietal cortex in the rhesus monkey. J. Comp. Neurol. 204, 196 –210. Rizzolatti, G., Luppino, G., Matelli, M., 1998. The organization of the cortical motor system: new concepts. Electroencephalogr. Clin. Neurophysiol. 106, 283–296. Rushworth, M.F.S., Nixon, P.D., Passingham, R.E., 1997. Parietal cortex and movement I. Movement selection and reaching. Exp. Brain Res. 117, 292–310. Sakata, H., Takaoka, Y., Kawarasaki, A., Shibutani, H., 1973. Somatosensory properties of neurons in the superior parietal cortex (area 5) of the rhesus monkey. Brain Res. 64, 85–102. Savaki, H.E., Dalezios, Y., 1999. 14C-Deoxyglucose mapping of the monkey brain during reaching to visual targets. Prog. Neurobiol. 58, 479 – 540. Savaki, H.E., Kennedy, C., Sokoloff, L., Mishkin, M., 1993. Visually guided reaching with the forelimb contralateral to a “blind” hemisphere: a metabolic mapping study in monkeys. J. Neurosci. 13 (7), 2772–2789. Smyrnis, N., Taira, M., Ashe, J., Georgopoulos, A.P., 1992. Motor cortical activity in a memorized delay task. Exp. Brain Res. 92, 139 –151. Snyder, L.H., Batista, A.P., Andersen, R.A., 1997. Coding of intention in the posterior parietal cortex. Nature 386, 167–170. Sokoloff, L., Reivich, M., Kennedy, C., Des Rosiers, M.H., Patlak, C.S., Pettigrew, K.S., Sakurada, O., Shinohara, M., 1977. The [14C]-deoxyglucose method for the measurement of local cerebral glucose utilization: theory, procedure, and normal values in the conscious and anesthetized albino rat. J. Neurochem. 28, 879 –916. Ungerleider, L.G., Desimone, R., 1986. Cortical projections of visual area MT in the macaque. J. Comp. Neurol. 248, 190 –222. Wise, S.P., 1985. The primate premotor cortex: past, present, and preparatory. Annu. Rev. Neurosci. 8, 1–19. Wise, S.P., Boussaoud, D., Johnson, P.B., Caminiti, R., 1997. Premotor and parietal cortex: corticortical connectivity and combinatorial computations. Annu. Rev. Neurosci. 20, 25– 42. Woolsey, C.N., 1958. Organization of somatic sensory and motor areas of the cerebral cortex, in: Harlow, H.F., Woolsey, C.N. (Eds.), Biological and Biochemical Bases of Behavior, Univ. Wisconsin Press, Madison, WI, pp. 63– 81.
