Topographic organization of baboon primary motor cortex: face, hand, forelimb, and shoulder representation.

Topographic Organization of Baboon Primary Motor Cortex: Face, Hand, Forelimb, and Shoulder Representation

Somatosens Mot Res Downloaded from informahealthcare.com by Queen's University on 01/01/15 For personal use only.

Robert S. Waters,* Donald D. Samulack,t Robert W.Dykes,t and Patricia A. McKinley$ *Department of Anatomy and Neurobiology, College of Medicine, University of Tennessee-Memphis, 875 Monroe Avenue, Memphis, Tennessee 38163; ?Department of Physiology, University of Montreal, Montreal, Quebec, Canada; and #Department of Physical and Occupational Therapy, McGill University, Montreal, Quebec, Canada Abstract ( I ) The fine details of the motor organization of the forelimb, face, and tongue representation of the baboon (Papio h. anubis) primary motor cortex were studied in four adult animals, using intracortical microstimulation (ICMS). (2) A total of 293 electrode penetrations were made. ICMS was delivered to 10,052 sites, and of these, 6,186 sites were verified to have been located within the grey matter. Motor effects were evoked from 30% of these sites. (3) The baboon motor cortex is confined, in large part, to the cortical tissue lying along the anterior bank of the central sulcus. When the electrode penetrations were confined to the precentral g y m , few sites were capable of evoking movement when stimulated by currents of 40 p A or less. (4) The details of the motor maps varied among the four animals; nonetheless, a general topographic organization existed, with the tongue musculature being represented most laterally, followed by a medial progression of the face, digits, wrist, forearm, and shoulder. Within the representation of a given body part, the muscles were organized as a mosaic, wherein the same muscle was multiply represented. (5) A zone of unresponsive cortex was observed to lie consistently between the face and forelimb representation in all four animals. Repeated electrode penetrations within the unresponsive zone failed to elicit muscle contractions even with stimulating currents as high as 80 pA. (6) Our results suggest that the baboon motor cortex is topographically organized; however, embedded within this overall pattern lies a fine-grained mosaic incorporating multiple representations of the same muscle. Key words primate, motor cortex, intracortical microstimulation, muscles,

EMG

Surface electrical stimulation of the pl'eoentral cortex has been used by a number of investigators to study the motor organization in human and nonhuman primates (see Humphrey, 1986, and Lemon, 1988, for reviews). Attempts to describe the organization of primary motor cortex were advanced significantly by the development of the intracortical microstimulation (ICMS) technique (Asanuma and Sakata, 1967). This method employs stimulating currents Somntosensory and

Motor Rcsesrcb, Vol. 7, No. 4.

1990. pp. 485-514

485


WATERS ET AL.

several hundred times lower than those used for surface stimulation. Using ICMS, Asanuma and Rosen (1972) concluded that the motor cortex was organized into discrete efferent zones, each specific for activating an individual muscle, and that the same muscle could be represented more than once in a pattern in which the efferent zone for an individual muscle had sharp boundaries and measured approximately 1 mm in diameter. They described this pattern of muscle efferent zones as being a mosaic. Data from other laboratories have not consistently supported the hypothesis of discrete efferent zones (Andersen et al., 1975; Jankowska et al., 1975a,b; Futami et al., 1979; Shinoda et al., 1981; Lemon et al., 1987), and the ICMS technique has not been without controversy (Jankowska et al., 1975a,b; McIlwain, 1982; Lemon et al., 1987). However, other investigators have used the ICMS method to map large portions of primary motor cortex (Kwan et al., 1978; Stick and Preston, 1978, 1982; McGuinness et al., 1980; Sessle and Wiesendanger, 1982; Huang et al., 1988) or its entirety (Gould et al., 1986), greatly enhancing our understanding of motor organization. In these experiments, visual observation of movements of body parts or muscle contractions were used to establish the relationship between the site of stimulation and the motor response. In all cases, the lowest stimulus current for activating a particular body part or muscle was defined as threshold. Through systematically stimulating a large expanse of cortex, investigators created maps of the representation of the body musculature in the motor cortex. Stick and Preston (1978) used ICMS and identified two separate representations of the digits and wrist in the motor cortex of the squirrel monkey. In the macaque monkey, Kwan et al. (1978) reported multiple representations of the forelimb organized around a single joint, which they described as a "nested" organization. In this configuration, the finger zone representation was surrounded by zones associated with more proximal joints. However, the movement of the same body part was often reported at several locations within a particular zone of the motor cortex. A similar pattern of organization was described in macaque monkeys by Sessle and Wiesendanger (1982), but investigators working with other species of monkeys were unable to confirm the nested pattern. McGuinness et al. (1980), Gould et al. (1986), and Huang et al. (1988) have reported a mosaic of muscle representations within a broader topographic order. Gould et al. (1986) mapped the motor representation of the entire body musculature in monkey, whereas McGuinness et al. (1980) and Huang et al. (1988) focused on mapping the face and tongue musculature in the monkey. Thus, the ICMS technique has revealed at least three different patterns of organization in primate motor cortex: dual, nested, and mosaic. The goals of the present study were to examine the details of the forelimb motor organization in the olive baboon (Pupio h. unubis), a species of monkey with a highly developed hand structure and large brain size, and to compare our results with those of other investigators in an effort to resolve the divergent views of motor organization. The results fiom our study suggest that in this species the primary motor cortex contains a clear somatotopic organization: The tongue musculature is represented most laterally, followed by a medial progression of face, hand, forearm, and shoulder musculature. The major cortical motor representation is located along the anterior bank of the central sulcus and extends only slightly onto the crown of the gyms. Within the representation of any body part, we found a mosaic arrangement where the same muscle could be rerepresented many times. A region of cortex unresponsive to ICMS was consistently found to lie between the face and hand 'k$xesentation. This Iatter observation confirmed an early report (Leyton and Sherrington, 'lV17) of unresponsive regions in the motor cortex of ape. 486

ORGANIZATION OF BABOON MOTOR CORTEX

MATERIALS AND METHODS Experiments were carried out in four adult baboons (Pupio h. unubis) weighing between 10 and 12 kg. Each was studied continuously for 108-132 hr.


ANIMAL. PREPARATION

The animals were tranquilized with a combination of ketamine (5.25 mgkg) and xylazine (0.45 mgkg). The hair was shaved on the head, neck, arm,and shoulder. Venous, arterial, and urinary catheters were installed, as well as a tracheal cannula and a rectal thermometer. Vital signs (body temperature, heart rate, and blood pressure) and electrocardiogram were monitored throughout the experiment. Measures of blood pressure levels, COz, p02, pH, Cl-,HC03-, Na', and K+ were obtained every 4 hr. Body temperature was maintained between 36.5" and 37.5"C by a circulating water blanket, supplemented when necessary by a heating lamp. Antibiotic agents, knlong S (1 1,OOO IU/kg/day procaine and benzathine penicillin G,i.m.) and netilmicin sulfate (50 mg, i.m.), were administered. Dextrose (5%) in 0.45% saline, supplemented with potassium, was infused through a venous catheter in the saphenous vein as required to compensate for fluid loss. SURGICAL. PREPARATION

Duriag inhalation anesthesia consisting of a mixme of 4096 nitrous oxide and 60% oxygen, supplemented with 0.5-2.096 halothane, the bone overlying the precentral sulcus was removed, the underlying dura was retracted, and dental impression compound was melted onto the bone surrounding the opening to form a retaining well. The surface of the cortex was covered with warmed Elliott's solution. During the mapping procedure, the Elliott's solution was replaced by a warmed heavy-grade silicon fluid (10,OOO cST) to reduce cardiovascular- and respiratory-induced movements and to provide thermal insulation. The cisterna magna was opened and allowed to drain, to further reduce the effects of cardiovascular and respiratory movements. A rod secured to the animal's skull with dental acrylic was co~ectedto a post fastened to the surgical table; this rod rigidly supported the head and upper torso, while simultaneously allowing free access to the face and mouth of the animal. Photographs of the cortical surface were used to record the entry site of all electrode penetrations. Following the surgical preparation, animals were maintained on inhalation anesthesia, occasionally supplemented with sodium pentobarbital (2- 10 mg/kg) as necessary to reduce spontaneous muscle twitches. IhTRACORTlCAL, MICROSTIMULATION

Tungsten-in-glassstimulatingmicroelectrodesmeasuring20-30 p m from the glass insulation to the tip were used (Stoney et al., 1%8). In an effort to sample the entice face, forelimb, and shoulder representations, penetrations were spaced approximately 1.O mm apart in the first two experiments.In the last two experiments,the penetrationswere placed closer together to produce a higher-resolution map of the forelimb, face, and the interveningcortical tehtory. 'Ihe microelectrode was advanced in 200-p steps, using a modified Canberra-type microdrive (Narishige). A single electrode penetration often exceeded 10 mm in length, allowing up to 487

WATERS ET AL.


50 intracortical sites to be tested per penetration. ICMS, consisting of a train of monophasic

cathodal pulses (12- 13pulses/train,0.2-msec pulse duration, 300 Hz,40-pA search stimulus intensity), was delivered to the cortical tissue through the microelectrode at a rate of 0.51.0 Hz. If a motor response was detected with the search stimulus, the current level was lowered until the response disappeared, and then gradually raised to the stimulus intensity that evoked a response on at least three consecutive stimulus presentations. This level was defined as threshold. When two or more muscles were activated at the same site, individual threshold measurements were made for each muscle. Stimulus intensities as high as 80 pA were used sparingly to test unresponsive locations. Current strength was continuously monitored on a storage oscilloscope as the voltage drop across a 1O-kfl resistor placed in series with the stimulating electrode. Motor responses from ICMS were detected (1) by visual observation; (2) by palpation of the muscle during contraction; and (3) occasionally by electromyograms (EMGS), particularly to explore unresponsive cortex (see below). A movement was recorded only after two observers agreed on the designation of the activated muscle@).The movements were described in terms of flexion, extension, adduction, abduction, medial and lateral rotation, and supination. The individual muscles inferred to be responsible for these movements were taken from the atlas by Swindler and Wood (1973). Each muscle group, identified with the three-letter abbreviation used throughout the text and figures, is shown in Table 1. The movement elicited was also recorded. Describing the muscles of the tongue proved to be the most difficult. ICMS affected small muscles on both the dorsal and ventral sides of the tongue, as well as ipsilateral and contralateral to the stimulating site in cortex. Within the dorsal and ventral locations, we partitioned the muscles into three compartments: anterior (tip), medial, or posterior. Therefore, an individual muscle within a single tongue compartment would have a designation that included its location in reference to the dorsal and ventral as well as the ipsilateral and contralateral sides (see Fig. 15 for details). The “Results” section describes regions within the motor representation from which ICMS does not appear to elicit motor responses. These regions are referred to as unresponsive zones or unresponsive cortex. In trying to eliminate technical or artifactual explanations for this observation, we had to rule out the possibility that these unresponsive zones were producing imperceptible movements. To test this hypothis, EMGs were recorded differentially from pairs of 25-pm Teflon-insulated silver wires. EMGs were amplified by standard electrophysiological equipment and recorded on tape. The insulation was removed 1 mm from the end and inserted into the belly of a muscle, using a hypodermic needle as a guide. Individual EMG leads were generally separated by less than 5 mm, particularly in smaller muscles. Groups of two to four muscles (the thenar muscles, some of the lumbricalismuscles, the period muscles, and the forearm muscles) were studied in this way.

HISTOLOGICAL RECONSTRUCTION At the end of selected electrode penetrations, electrolytic lesions were ma& by passing anodal current of 10 pA for 10-20 sec. Prior to perfusion, animals were deeply anesthetized with Nembutal(40 mg/kg), and the boundaries of the recording area were marked by inserting tungsten wires into the cortex at the same orientation used to make the electrode penetrations. The tips of fine glass micropipettes were also inserted into the cortex and broken off at selected sites to aid in the reconstruction. The animals were perfused with 0.9% saline and 488



TABLE1. Muscle Abbreviations Abbreviation ADM ADP APB APL BUC CHT DAO DEL ECU EM3 EDP EPL EQP FDP FDS FPB IOD IOP ISP

LAO WM NLB ODM

OOR OPP PMA TCL TPL

TPZ VIB ZYG

To4 T40

Muscle Abductor digiti minimi Adductor pollicis Abductor pollicis brevis Abductor pollicis longus Buccinator Contrahentes Depressor anguli oris Deltoideus Extensor carpi ulnaris Extensor digitorum communis Extensor digiti secundi et tertii proprius Extensor pollicis longus Extensor digiti quarti et quinti proprius Flexor digitorum profundus Flexor digitorum superficialis Flexor pollicis brevis lnterossei dorsales Interossei palmares Infraspinatus Levator anguli oris Lumbricales Nasolabialis Oppomns digiti minimi Orbicularis oris Opponens pollicis Pectoralis major Triceps brachii caput longum Platysma (traeneloplatysma) Trapezius Vibrissae Zygomaticus Tongue (dorsal, medial, contralated) Tongue (venml. medial, conhalateral)

10% buffered formalin, and the brains were removed. The region containing the central sulcus was blocked and prepared for sectioning. Serial frozen sections were cut from the first brain at W p m thickness and stained with thionine. The other three brains were embedded in celloidin and sectioned at either 80- or 120-pm steps. Every fifth section was drawn with the aid of a projection microscope and used to reconstruct the precentral gyrus and anterior bank of the precentral sulcus. First, the serial sections were transposed onto a single page, where they were arranged in order to reconstruct the wall of the central sulcus as well as the mown of the precentral gyrus. Then each penetration was located in the sections on the basis of lesions, marking wires or glass, and vascular landmarks recorded on photographs taken of the surface at the beginning of the experiment. These identifying aids were used to determine which parts of the precentral gyrus were capable of eliciting movements, and this information was transposed to the appropriate section of the reconstruction. In addition, many of the penetrations were reconstructed and superimposed onto line drawings of the 489


WATERS ET AL.

respective section(s) in which they were located, so that the data could be related to the details of cytoarchitecture and cortical laminae found in that section. A second approach to representing the data was to assign an X and Y coordinate in millimeters to each penetration, and the depth of each stimulating site became a Z coordinate for a three-dimensional matrix of data points. These coordinates were stored in a computer with information about the activated muscle, the threshold, the threshold for the second muscle (if present), and its identity. The final entry on each line was the name of the body part affected by the movement observed. Referring to the information from the histological reconstructions enabled us to remove any data point located in white matter. The remaining set, consisting of data points identified in the grey matter, was manipulated by a commercial plotting package (Autocad) that allowed the data to be plotted in a simulated three-dimensional array; this permitted us to examine the spatial relationships among the data points. RESULTS

DATA SET

A total of 293 penetrations were made in four animals. The penetrations near the sulcus often extended for a distance of 10 mm, allowing as many as 50 distinct measurements to be made along a single electrode track down the bank of the sulcus. In all, 10,052 sites were explored using microstimulation. Table 2 summarizes the data set obtained. Out of a total of 6,186 sites (61.5%) stimulated in grey matter, we were able to elicit movements in 1,849 (30%).Ninety-one percent (n = 1690) could be activated by stimulating currents of 40 PA or less (Fig. 1). GENERAL ORGANIZATION OF BABOON MOTOR CORTM

Figure 2 shows the general location of the tongue, face, and forelimb portions of the motor representation for three of the animals studied. (The map of the fourth animal was nearly identical to the map shown in Fig. 2A.). This figure represents lowest-threshold sites for muscle contraction, which may or may not be accompanied by movement of a body part, which we refer to as muscle movement. The area of explored cortex extended nearly 20 mm in the mediolateral direction, and antemposteriorly it included both the bank of the TABLE2. Characteristics of the Data Set Obtained from Four Baboons Animal

No. penetrations No. with movement Rrcentage

No. sites tested No. sites in grey matter No. sites eliciting movement in grey matter Percentage 490

Total

1

2

3

4

80 39 49

67 31 46

106 47 44

44 64

297 145 49

1945 1155

2712 1442

3565 1835

1830 1755

10.052 6186

490 42

528 37

so2

329 19

1849

27

28

30


T


25

n = 1849

T

0

110

$20

$30

$40

$50

160

>60

THRESHOLD CURRENT ( PA)

FIGURE1 . Distribution of threshold currents shown as a percentage of the total sample for all sites from which responses were elicited in the grey matter.

precentral sulcus and the precentral gyrus. However, in each animal, the majority of cortex from which movements could be elicited was localized to the sulcal bank, while only a limited portion of the representation extended onto the surface of the gyrus. Since this finding had been unexpected, the crown of each gytus was carefully studied with multiple penetrations, but no motor responses were found. The body representation began most laterally with a large area devoted to the tongue, which overlapped with the immediately adjacent face representation. Medial to the cortex activating muscles of the face was a large expanse of cortex, extending to the depth of the central sulcus from which it was impossible to elicit muscle contractions even when the stimulatingcurrent was elevated to 80 FA. This unresponsivearea (discussed in detail below) waa bounded on the medial side by cortex, where microstimulationtriggered movements of the intrinsic hand muscles. The hand representation was usually contiguous with cortex where microstimulation was capable of producing movements in the forearm muscles. The shoulder representation was found medial to the forearm representation. Although this general pattern of motor organization was seen in each animal, it can be seen from Figure 2 that there were individual differences in size, shape, and location of responsive and unresponsive zones of cortex. Figure 3A illustrates a camera lucida reconstruction of four electrode penetrations located in a single parasagittal section passing through the hand representation of one animal. The majority of the sites capable of eliciting movements were localized in the bank of the precentral d c u s and on the convexity of the gyrus. 'Ihe two most antedor electrode penetrations (3, 4) failed to elicit movement even with stimulating currents as high as 60 PA. In the other two penetrations (1, 2), ICMS produced contractions of the muscles controlling the 6rst digit (Dl). As the electrode traversed the gyral convexity and sulcal bank, it activated 49 1


WATERS

ET AL.

C

FIGURE 2. Reconstruction of the topographical organization of body part representation (tongue, face, digits, wrist. elbow, shoulder) for three of the animals studied, as viewed from an observation point within the central sulcus, looking in an anterior direction toward the precentral region. (The fourth animal was nearly identical to the case illustrated in A.) The reconstruction was generated by transferring the data concerning the activated body part in each electrode penetration radially onto the cortical surface of each reconstructed section. The sections were then aligned to produce the map or the threedimensional image. T,tongue; F, face; D,digit; E,elbow; s, shoulder; W,wrist. Arrows on the left indicate that additional areas of the representation were found further laterally.

muscles at cortical sites with stimulating currents between 5 p A and 50 pA. Muscle contrsactions were elicited from seven different muscle groups at different points along the penetrations. Overlap between muscle groups occurred three times. In two of these cases, the region of overlap was restricted to a distance of less than 0.5 mm. In the second penetration, muscle contractions were elicited in the white matter and within the bank of the sulcus, illustrating that motor responses were often elicited from within the white matter; in a few of these 492



cases, threshold was as low as 5 PA. This finding indicates that microstimulation is effective in eliciting motor responses from myelinated axons as well as from cell bodies. Figure 3B is a comparable reconstruction of a histological section passing through the shoulder representaton of the same animal, in which five electrode penetrations were identified. In this example, the muscle repmentation also was limited primarily to the bank of the sulcus and gyral convexity. Penetration 5 encountered unresponsive corkx on the crown of the gyrus; however, in penetration 4, ICMS activated the triceps muscle from within the gyms.This penetration may be compared to penetration 3, which yielded a motor response only deep in the sulcal wall. Penetrations 1 and 2 were successful in activating muscles of the upper arm and shoulder from sites within the gyral convexity and sulcus. Here ICMS produced muscle contraction in the pectoral muscles. Again, as was the case in the part of the representation serving the hand shown in Figure 3A, the cortex capable of causing muscle contractions did not extend very far onto the gyral surface, but rather was limited to the convexity and wall of the gyrus.

U Imr,

FIGURE3. Recoostmaion of electrode penetrations through the digit and shoulder repsentation of motor cortex. (A) Parasagittal sectioo showing the locatioo of four recoostructed electrode tracks; in two of these penetrations, responses were elicited from muscles controlling digits. 10 one penetration where motor effects were obtained, two instances were observed where two muscle contractions occurred at the same threshold at a single location. (B) Five electrode penetrations are reconstructed in a parasagittal sectioo, and in four of these penetrations, motor responses were observed in the shoulder representation. In two penetrations, more than one muscle response could be activated at threshold. 493


WATERS

ET AL.

Figure 4 shows the same reconstructions for the responsive penetrations illustrated in Figure 3, and the data concerning the individual muscle groups have been replaced by information about the lowest currents that could activate the identified muscle groups. The first point to note is that the different muscle groups encountered along the penetration appeared rather abruptly. A new muscle became active within a distance of 250-500 pm; except for several places in penetrations 1 and 2 of Figure 4A, where two muscles could be activated simultaneously, the effective region for each muscle seemed to have a rather clear border with the region serving the next muscle. Penetration I , in Figure 4A, illustratesthe well-establishedobservationthat the threshold currents required to elicit movements are higher in the superficial layers of cortex. As the electrode reached the middle layers of cortex, the movements of the muscle could be elicited by much lower currents. Then in this penetration, ICMS began to elicit movements of two muscles, the flexor pollicis brevis (FPB)and interosseus dorsalis (IOD), for a distance of 0.5 nun. bllowing this, at appxirnately 0.5-mm intervals new muscle groups were ellcOuntered abruptly at a threshold below 20 pA and sometimes as low as 2 pA. In penetration 2, no muscles were activated until the electrode reached a depth of 4.7 mm below the cortical surface. At this point deep within the white matter, ICMS elicited movements from a region conveying axons to and from the cortex in the wall of the sulcus. Generally the thresholds were high in the white matter, but still the motor responses changed

1

2

B

U 1-

FIGURE4. Same reconstructions as illustrated in Figures 3A and 3B, but for those sites producing motor effects, the threshold currents for activating each muscle movement are plotted. 494



abruptly to a different muscle as the electrode was advanced. As the electrode entered layer VI of the grey matter, the threshold decreased. For 200 pm the extensor pollicis longus (EPL) and FPB were activated at the same threshold; then the FPB response disappeared, and the threshold for the EPL dropped to 2 p A and stayed low for 1.5 mm. Finally, the responses changed to the opponens pollicis (OPP), and the EPL threshold increased to 25 pA within a distance of 200 pm. Figure 4B illustrates the threshold data for the three penetrations shown in Figure 3B passing through the shoulder representation. In the anterior two penetrations, the threshold for muscle movement became progressively more elevated on one or both sides of the location that most effectively activated a muscle, but this did not happen in penetration 1. Penetration 1 illustrates another example of a penetration where different muscle groups were encounted sequentiallyalong the penetration, with rather abrupt, low-threshold transitions between adjacent sites activating different muscles.

ICMS THRESHOLDS Threshold currents for a particular muscle were examined to determine whether there was a systematic variation of threshold associated with body part served or with cortical layer stimulated. Figure 5 illustrates the results of an analysis of a subset of the data in the four animals, where the stimulationsite could be assigned to a specific cortical layer in an identified electrode penetration. These criteria excluded sites on the borders of two cortical layerssites where there might be an error of scale, and sites in histological sections that were inadequately stained to permit us to identify the laminar boundaries-leaving a total of 209 sites from the four animals. Cortical layers II and III were combined. Layers 11-III and layer VI were each divided into four bins. For the thinner layers, I and V, all data were placed in a single bin. Threshold measurements were assigned to one of these four bins

Mean-

(MI

FIGURE5 . The relationship between the distribution of mean thresholds and cortical layer. Cortical layers Il and IlI are combined and partitioned into four parts, as was the wide layer VI. 495


WATERS ET AL.

according to which portion of the layer the electrode was in when the measurement was obtained. Low thresholds were found between layer 111 and the superficial portion of layer VI. Note particularly that the lowest thresholds were found in the part of lamina VI immediately beneath layer V. As expected, the highest thresholds were found in layer I and the deeper portion of layer VI. Stimulations sites within the white matter required higher intensities to elicit muscle contractions than did stimulation sites in the middle cortical layers, but the average threshold in the white matter was comparable to that found in layer I. Nonetheless, thresholds of less than 20 pA were not uncommon for sites within the white matter. The observation that the lowest thresholds were found just below layer V was unexpected, and led to a reexamination of individual electrode penetrations that had been carefully reconstructed to show the relationship between the cortical layers and the threshold data. Although low-threshold points were found in most layers, this reexamination confirmed that the upper part of layer VI most consistently produced low-threshold sites. Figure 6 illustrates this phenomenon with a fortuitous penetration in the face representation that traversed the upper part of layer VI for several millimeters. In this region, several different muscles were encountered, each having a threshold of less than 3 pA.

A

B

FIGURE6. Reconstruction of electrode penetration in the tongudface representation, which traversed a large portion of the upper pm of layer VI. (A) A parasagittal section showing the reconstruction of a single electrode tract and the muscles activated along the tract. (B) The same electrode tract showing the relationship between the stimulating site where the motor response was elicited and the threshold recorded at that site. 496


FACE

TONGUE

I

13.4


"

0

5 10152025303540455055606570

1-134

" 0

5 10152025303540455055606570

DIGITS

WRIST

I

n.4 16

n-4

16'

12 %

8 4 n

v

0 5 10152025303540455055606570

"

0 5 10152025303540455055606570 PA

PA

THRESHOLDS FOR MOVEMENT

FIGURE7. Illustration of the relationship between threshold of response and percentage of the total cases observed for representations of

the tongue, face, wrist, and digits.

The threshold data were also divided into categories according to the body part served by the muscle activated. The data from all four animals were combined, and the probability of encountering a particular threshold was expressed as a percentage of all threshold measurements made for muscles serving a particular body part. Figure 7 illustrates the results of this comparison for the face, tongue, wrist, and digits. It is clear that the largest percentages of low-threshold sites (

Primary somatosensory cortex hand representation dynamically modulated by motor output.

Double representation of the distal forelimb in cat motor cortex.

Organization of the forelimb area in squirrel monkey motor cortex: representation of digit, wrist, and elbow muscles.

The origin of thalamic inputs to the "hand" representation in the primary motor cortex.

Functional organization of the avian auditory cortex analogue. I. Topographic representation of isointensity bandwidth.

Somatosensory cortex of the neonatal pig: I. Topographic organization of the primary somatosensory cortex (SI).

The topographic organization of rhesus monkey prestriate cortex.

Impairments in reaching during reversible inactivation of the distal forelimb representation of the motor cortex in the cat.

Topographic organization of afferent inputs in monkey precentral cortex.

Multiple representation in the primate motor cortex.

Functional topographic organization of the motor reticulothalamic pathway.

Topographic distribution of connections from the primary motor cortex to the corpus striatum in Aotus trivirgatus.

Forelimb training drives transient map reorganization in ipsilateral motor cortex.

Neuronal representation of stand and squat in the primary motor cortex of monkeys.

Primary Motor Cortex Representation of Handgrip Muscles in Patients with Leprosy.

Somatosensory cortex of the neonatal pig: II. Topographic organization of the secondary somatosensory cortex (SII).

Neural Representation and Causal Models in Motor Cortex.

Topographic NMF for data representation.

Neurophysiological correlates of hand preference in primary motor cortex of adult squirrel monkeys.

Prediction of hand trajectory from electrocorticography signals in primary motor cortex.

Motor Recovery of the Affected Hand in Subacute Stroke Correlates with Changes of Contralesional Cortical Hand Motor Representation.

Predicting hand orientation in reach-to-grasp tasks using neural activities from primary motor cortex.

The laryngeal motor cortex: its organization and connectivity.

Topographic organization of the basal forebrain projections to the perirhinal, postrhinal, and entorhinal cortex in rats.