Properties of kinesthetic neurons in somatosensory cortex of awake monkeys

(1) To study neural mechanisms used to encode kinesthetic information in somatosensory cortex of awake monkeys, we recorded from 227 single neurons responsive to joint movement or specific postures of the forelimb or hand (kinesthetic neurons). Unit responses were characterized quantitatively with respect to: (a) firing patterns; (b) responses to ramp changes in joint position and joint velocity; and (c) responses to sinusoidal joint movements.

(2) Kinesthetic neurons were divided into 3 groups. Rapidly-adapting neurons (44%) responded only to joint movement, giving a burst of impulses proportional to velocity. They showed no tonic responses to limb posture. Two populations of tonically active neurons were observed: slowly-adapting neurons (43%) and postural neurons (13%). Both types increased their firing rates with increasing degrees of flexion or extension, showing maximum excitation at the extremes of joint position in the preferred direction. They were distinguished by their sensitivity to the velocity of movement, the size of the angle over which they respond, and the phase relation of their responses to sinusoidal joint movement.

(3) The firing rates of kinesthetic neurons in S-I cortex are functions of both joint angle and joint velocity. The importance of each component varies in the 3 classes: velocity of movement is the most important determinant of firing rates of rapidly-adapting and slowly-adapting kinesthetic neurons, and joint angle predominates the responses of postural neurons.

Receptive fields of neurons in areas 3b and 1 of somatosensory cortex in monkeys

Receptive fields of neurons within the separate representations of the glabrous hand in areas 3b and 1 of somatosensory cortex were studied in cynomolgus monkeys. Many neurons in area 1 have center-surround receptive fields with separate 'on' and 'off' zones, while neurons in area 3b exhibit largely uniform or homogeneous receptive fields.     

Recognition of temporally structured activity in spontaneously discharging neurons in the somatosensory cortex in waking cats

We describe a method to automate the detection and analysis of structured neuronal activity obtained in relatively non-restrictive experiments in awake animals. Several different, regularly occurring, discharge patterns consisting of groups of spikes were identified in extracellular recordings from the somatosensory cortex of awake cats. The introduction of an interspike interval threshold made it possible to segregate these bursts from single spikes. The threshold interval was obtained from the modal interval in high-resolution autocorrelograms (up to 0.1 ms/bin) of the spontaneous neural activity. Single spikes were those separated by intervals greater than the threshold, while those within the group were of less than threshold value. When intervals were arranged and averaged according to their order of occurrence within the burst, four distinctive burst patterns were observed. These four patterns occurred in both normal and deafferented cortex and we believe them to be characteristic of particular cell types, a feature that will be useful for studying such cells in intact cellular networks.     

Representations of the body surface in areas 3b and 1 of postcentral parietal cortex of cebus monkeys

The somatotopic organization of postcentral parietal cortex was determined with microelectrode mapping methods in a New World monkey, Cebus albifrons. As in previous studies in macaque, squirrel and owl monkeys, two separate representations of the body surface were found in regions corresponding to the architectonic fields 3b and 1. The two representations were roughly mirror-images of each other, with receptive field locations matched for recording sites along the common border. As in other monkeys, the glabrous digit tips of the hand and foot pointed rostrally in the Area 3b representation and caudally in the Area 1 representation. Both representations proceeded in parallel from the tail on the medial wall of the cerebral hemisphere to the teeth and tongue in lateral cortex along the Sylvian fissure. Compared with the other monkeys, the tail of the cebus monkey, which is prehensile, was represented in a very large region of cortex in Areas 3b and 1. Like its close relative, the squirrel monkey, the representation of the trunk and parts of the limbs were reversed in orientation in both Area 3b and Area 1 in cebus monkeys as compared to owl and macaque monkeys. The reversals of organization for some but not all parts of the representations in cebus and squirrel monkeys suggest that one line of New World monkeys acquired a unique but functionally adequate pattern of somatotopic organization for the two adjoining fields.     

Cortical connections of the second somatosensory area and the parietal ventral area in macaque monkeys

To gain insight into how cortical fields process somatic inputs and ultimately contribute to complex abilities such as tactile object perception, we examined the pattern of connections of two areas in the lateral sulcus of macaque monkeys: the second somatosensory area (S2), and the parietal ventral area (PV). Neuroanatomical tracers were injected into electrophysiologically and/or architectonically defined locations, and labeled cell bodies were identified in cortex ipsilateral and contralateral to the injection site. Transported tracer was related to architectonically defined boundaries so that the full complement of connections of S2 and PV could be appreciated. Our results indicate that S2 is densely interconnected with the primary somatosensory area (3b), PV, and area 7b of the ipsilateral hemisphere, and with S2, 7b, and 3b in the opposite hemisphere. PV is interconnected with areas 3b and 7b, with the parietal rostroventral area, premotor cortex, posterior parietal cortex, and with the medial auditory belt areas. Contralateral connections were restricted to PV in the opposite hemisphere. These data indicate that S2 and PV have unique and overlapping patterns of connections, and that they comprise part of a network that processes both cutaneous and proprioceptive inputs necessary for tactile discrimination and recognition. Although more data are needed, these patterns of interconnections of cortical fields and thalamic nuclei suggest that the somatosensory system may not be segregated into two separate streams of information processing, as has been hypothesized for the visual system. Rather, some fields may be involved in a variety of functions that require motor and sensory integration.     

Interactions among convergent inputs to somatosensory cortex neurons

Postsynaptic potentials (psps) produced by electrical stimulation of 4 forelimb nerves were recorded intracellularly from neurons in the primary somatosensory cortex of sodium pentobarbital anesthetized cats. Convergent inputs were found from nerves subserving different modalities and different regions of the forelimb. Psps from separate afferent sources usually did not sum linearly but rather interacted with one another. These interactions could have occurred at the cortical level or earlier in the ascending pathways and are interpreted with regards to the control of somatosensory responsiveness by multiple converging inputs.     

Timing and connectivity in the human somatosensory cortex from single trial mass electrical activity

Parallel-distributed processing is ubiquitous in the brain but often ignored by experimental designs and methods of analysis, which presuppose sequential and stereotypical brain activations. We introduce here a methodology that can effectively deal with sequential and distributed activity. Regional brain activations elicited by electrical median nerve stimulation are identified in tomographic estimates extracted from single trial magnetoencephalographic signals. Habituation is identified in both primary somatosensory cortex (SI) and secondary somatosensory cortex (SII), often interrupted by resurgence of strong activations. Pattern analysis is used to identify single trials with homogeneous regional brain activations. Common activity patterns with well-defined connectivity are identified within each homogeneous group of single trials across the subjects studied. On the contralateral side one encounters distinct sets of single trials following identical stimuli. We observe in one set of trials sequential activation from SI to SII and insula with onset of SII at 60 msec, whereas in the other set simultaneous early co-activations of the same two areas.     

Pyramidal tract neurons in somatosensory cortex: central and peripheral inputs during voluntary movement

Recordings from pyramidal tract neurons (PTNs) in the primary somatosensory cortex of the monkey show that these neurons have 3 properties in common with PTNs of primary motor cortex: (1) they exhibit discharge prior to the onset of voluntary movement, (2) their discharge frequency varies as a function of strength of muscular contraction, and (3) they show reflex responses to afferent stimuli that occur during movement. These findings support the view that in addition to its widely recognized role in somesthetic perception, somatosensory cortex has a direct role in the control of movement.     

The effects of peripheral deafferentation on spontaneously bursting neurons in the somatosensory cortex of waking cats

Single neurons (n=356) were studied in the forelimb representation of awake, quietly resting cats. Thirty-five spontaneously bursting neurons in a sample of 206 cells recorded before forelimb deafferentation were compared to 39 spontaneously bursting neurons in a sample of 127 neurons studied 1–3 weeks after deafferentation. The probability of encountering bursting neurons increased significantly following deafferentation from 17% to 31% of the sample (P<0.005). The same 5 classes of bursting cells were observed after deafferentation but there were significant changes in the duration of interspike intervals in some classes, in the probability of observing certain classes, and in the proportion of spikes found in bursts. The probability of encountering class III cells, a class thought to consist primarily of non-inactivating pyramidal burst neurons, nearly doubled and the average interspike interval length within the burst increased from 1.9 to 3.0 ms. The burst structure in the other classes did not change but they were found less frequently. These other classes may include inhibitory interneurons which receive less excitatory drive after deafferentation and therefore provide less inhibition to class III cells. The differential behavior of the different classes of bursting cells may be one reason why the overall level of spontaneous activity does not change after deafferentation and it suggests that there are homeostatic mechanisms in primary somatosensory cortex that maintain a certain level of neural activity.     

Body surface maps in the somatosensory cortex of rabbit

The organization of somatosensory maps was examined in rabbits with the aid of microelectrode multi-unit recording techniques. Two complete maps of the contralateral body surface are identified in the parietal cortex. The first map, S I, is found entirely on the lateral convexity of the hemisphere and closely resembles S I described in the rat (Welker, '71, '76). It is organized in a complex, though systematic, fashion with the representations of the hindlimb and tail located caudomedially. These representations are followed laterally in sequence by those of the trunk and forelimb and then the representation of the head. Within the head representation the lips are found rostrally, the vibrissae caudomedially, and the displaced representation of the pinna of the ear is located caudolaterally. Unlike the disposition in most other mammals, the dorsal midline of the trunk is represented along the caudal border of S I. Within S I, the representations of the circumoral surfaces, including the lips, philtrum, nose, and vibrissae, are emphasized, occupying approximately 86.4% of the map. It is suggested that S I is contained within a single major koniocortical region, here called the medial parietal area, or Pm. The several previously described parietal regions (Rose, '31; Fleischhauer et al., '80) are interpreted as subregions that are related to particular representations of portions of the body surface. The second map, S II, is located lateral to S I in a region here called the lateral parietal area or Pl. S II shares a common border with S I along the representations of the philtrum, bridge of the nose, and top of the head. The body is oriented in an erect conformation with the head located rostrally and medially and the hindlimb and tail located caudally and laterally.     

Studies on the evolution of multiple somatosensory representations in primates: the organization of anterior parietal cortex in the New World Callitrichid, Saguinus

Because members of the New World family, Callithricidae, are generally regarded as the most primitive of monkeys, we studied the organization of somatosensory cortex in the tamarin (Saguinus) in hopes of better understanding differences in the organization of anterior parietal cortex in primates and how these differences relate to phylogeny. In most prosimian primates only one complete representation of cutaneous receptors has been found in the region of primary cortex, S-I, while in all Old and New World monkeys studied to date, two cutaneous representations exist in distinct architectonic fields, areas 3b and 1. In detailed microelectrode mapping studies in anesthetized tamarins, only one complete representation responsive to low-threshold cutaneous stimulation was evident in the S-I region. This topographic representation was in a parietal koniocortical field that architectonically resembles area 3b of other monkeys, and the general somatotopic organization of the field was similar to that of area 3b of other monkeys. Cortex rostral to the single representation was generally unresponsive to somatosensory stimuli, or required more intense stimulation for neural activation. Cortex caudal to the representation, in the region of area 1 of other monkeys, was generally either unresponsive or responded to only high-threshold stimulation, although some recording sites were activated by low-threshold tactile stimulation. The present evidence, together with that from previous studies, suggests that the single, complete body surface representation in Saguinus is homologous to the S-I representation found in some prosimians (Galago, Perodicticus) and the area 3b cutaneous representation found in New World Cebidae (Aotus, Saimiri, and Cebus) and Old World Macaca. Cortex rostral to S-I in Saguinus has the appearance of areas 3a and 4 of other primates. The cortex caudal to S-I in Saguinus, while resembling area 1 in some ways, does not have all of the features of area 1 of other monkeys. In particular, the field was not easily activated by low-threshold cutaneous stimuli, as area 1 is in other monkeys, and therefore a second cutaneous representation of all body parts was not demonstrated. Thus, cortex in the expected location of area 1 of Saguinus was not as responsive as area 1 of other monkeys, and it somewhat resembled the high-threshold fringe zones found caudal to S-I in anesthetized prosimians and some nonprimates. The results raise the possibility that the area 1 cutaneous representation that is characteristic of other New World monkeys and Old World monkeys evolved from a less responsive precursor along the caudal border of S-I in early monkeys.     

Connections of area 2 of somatosensory cortex with the anterior pulvinar and subdivisions of the ventroposterior complex in macaque monkeys

The principal goal of the present study was to determine the thalamic connections of area 2 of postcentral somatosensory cortex of monkeys. The placement of injections of anatomical tracers (horseradish peroxidase, wheat germ agglutinin, or ³H-proline) was guided by extensive microelectrode maps of cortex in the region of the injection site. These maps identified the body parts represented in the cortex included in the injection site, and provided information about the physiological boundaries of area 2, which was related later to the cortical architecture. Most injections were placed in the representation of the hand in area 2, which was highly responsive to cutaneous stimuli and could be mapped in detail. Injections were also placed in other parts of area 2, area 1, or area 5, and some injections involved more than one area. As other investigators have determined, regions of retrograde and anterograde thalamic label overlapped, demonstrating that connections with cortex are reciprocal. Injections completely confined to area 2 consistently produced label in two locations: the anterior pulvinar (Pa) and a dorsal capping zone of the ventroposterior complex that we term the ventroposterior superior nucleus (VPS). Single restricted injection sites resulted in one region of label in VPS, and multiple foci of label in Pa. In some cases where the injection was confined to the representation of the hand in area 2, label was also found more ventrally in the ventroposterior complex in ventroposterior nucleus proper (VP). Thus, area 2 receives input from Pa, VPS, and, at least in some locations and individuals, VP. Injections of tracers into area 1 confirmed previous findings that area 1 is densely interconnected with VP. In addition, there appear to be sparse connections with VPS. There was no evidence of connections with Pa. Evidence from injection sites that extended from area 2 into areas 5 and 7, and from injection sites in area 5, indicates that the lateral posterior nucleus (LP) projects to rostral areas 5 and 7. The results support the conclusion that area 2 is a functionally distinct subdivision of somatosensory cortex, and indicate that area 2 has thalamic connections that are characteristic of both “sensory” (VP and VPS) and “association” (Pa) cortical fields.     

Inhibitory modulation of cat somatosensory cortex: a pharmacological study

In anesthetized preparations, GABA and taurine produced rapid, reversible inhibition of the negative component (N20) of the primary somatosensory evoked potential (SEP) without effect on the earlier positivity (P11). This effect was also produced by low doses of 4-aminopyridine. Neither bicuculline or picrotoxin antagonized these drug effects. A predominance of type B GABA receptors in the superficial layers of the somatosensory cortex is proposed.     

Cooperative processing in primary somatosensory cortex and posterior parietal cortex during tactile working memory

In the present study, causal roles of both the primary somatosensory cortex (SI) and the posterior parietal cortex (PPC) were investigated in a tactile unimodal working memory (WM) task. Individual magnetic resonance imaging‐based single‐pulse transcranial magnetic stimulation (spTMS) was applied, respectively, to the left SI (ipsilateral to tactile stimuli), right SI (contralateral to tactile stimuli) and right PPC (contralateral to tactile stimuli), while human participants were performing a tactile‐tactile unimodal delayed matching‐to‐sample task. The time points of spTMS were 300, 600 and 900 ms after the onset of the tactile sample stimulus (duration: 200 ms). Compared with ipsilateral SI, application of spTMS over either contralateral SI or contralateral PPC at those time points significantly impaired the accuracy of task performance. Meanwhile, the deterioration in accuracy did not vary with the stimulating time points. Together, these results indicate that the tactile information is processed cooperatively by SI and PPC in the same hemisphere, starting from the early delay of the tactile unimodal WM task. This pattern of processing of tactile information is different from the pattern in tactile‐visual cross‐modal WM. In a tactile‐visual cross‐modal WM task, SI and PPC contribute to the processing sequentially, suggesting a process of sensory information transfer during the early delay between modalities.     

Short communication: mapping of somatosensory cortices with functional magnetic resonance imaging in anaesthetized macaque monkeys

Functional magnetic resonance imaging (fMRI) in macaque monkeys is emerging as a potent candidate to bridge the gap between data from human fMRI studies and data from anatomy, electrophysiology and lesion studies in monkeys. The primary (SI) and secondary (SII) somatosensory cortices are the principal regions for somatosensory information processing and contain systematic representations of the body surface map (somatotopy). To examine the functional organization of the somatosensory cortices in anaesthetized macaque monkeys with fMRI, we asked whether focal and differential activation could be observed in SI and SII in response to tactile stimulation with two parameters: body sides (right and left) and body regions (hand and face). We found that changes in stimulus parameters elicited differential focal activation in both SI and SII in two ways. First, the hand and face stimulation activated SI and SII in the contralateral, but not in the ipsilateral, hemisphere. Second, the hand and face stimulation differentially activated two adjacent regions in both SI and SII. These fMRI results appear to correlate with previous mapping studies by other methods in the macaque somatosensory cortices. This study shows the feasibility of fMRI studies in mapping multiple sensory areas in monkeys by which we can distinguish between adjacent functionally distinct regions.     

Effect of methamphetamine on glutamate-positive neurons in the adult and developing rat somatosensory cortex

The neurotoxic effects of methamphetamine (MA) on dopaminergic and serotonergic terminals have been well-documented. Another neurotoxic effect of MA is neuronal degeneration in the somatosensory cortex, as seen by silver staining. The neurochemical characteristics of these degenerating neurons are unknown. Using glutamate and glial fibrillary acid protein (GFAP) immunohistochemistry, it was found that MA exposure in adult rats (10 mg/kg given 4 times intraperotoneally(i.p.) at 2-h intervals) causes localized depletion of glutamate-positive neurons and astrogliosis in the somatosensory cortex 3 days following treatment. The affected region covered the middle one-third portion from the longitudinal fissure to the rhinal sulcus and was predominately seen in layers II-III of the cortex. This pattern of depletion is consistent with that demonstrated previously with silver staining following MA, d-amphetamine, and 3,4-methylenedioxymethamphetmine (MDMA) exposures. Comparable effects were not found in developing animals at ages previously shown to also be resistant to MA-induced effects on dopaminergic terminals (age 20 and 40 days). Results suggest that MA exposure induces degeneration of glutamatergic neurons in the somatosensory cortex of adult rats. © 1996 Wiley-Liss, Inc.     

Connectional networks within the orbital and medial prefrontal cortex of macaque monkeys

The intrinsic cortico-cortical connections within the orbital and medial prefrontal cortex (OMPFC) were demonstrated with retrograde and anterograde tracers injected into each of the architectonic areas that constitute this region. Although many of the connections linked neighboring areas, others selectively connected relatively distant areas. Most, but not all, of the connections were reciprocal. Altogether, the connections formed at least two distinct networks within the OMPFC. The “orbital” prefrontal network linked most of the areas within the orbital cortex, with very few connections to medial prefrontal areas. Areas Iam, Iapm, Ial, 121, 12m, and 12r in the caudal and lateral parts of the orbital cortex (which received inputs from several sensory modalities) had convergent connections with areas 13l, 13m, and 13b in the central orbital cortex, with further connections to the rostral orbital area 11l. For the connections between areas Iapm, Iam, Ial, 13m, 13l, and 11l, rostrally directed fibers arose mainly in layer V, whereas caudally directed fibers originated mainly in layer III. The “medial” prefrontal network selectively involved medial areas 14r, 14c, 24, 25, 32, and 10m, rostral orbital areas 10o and 11m, and agranular insular area Iai in the posterior orbital cortex. Two orbital areas, 13a and 12o, had substantial connections to both networks and may serve as points of interaction between them; otherwise there were relatively few interconnections. The two networks also had distinct connections with other cortical regions, with limbic structures, and with the mediodorsal thalamic nucleus. Their role in guidance of affective behavior is discussed. © 1996 Wiley-Liss, Inc.     

Corticothalamic connections from the second somatosensory area and neighboring regions in the lateral sulcus of macaque monkeys

Corticothalamic connections were shown between the second somatosensory area in primates and the ventroposterior nuclei of the thalamus. These projections were topographically arranged with those from the hindlimb portions of SII traced to the most lateral and posterior parts of the ventroposterior lateral nucleus (VPLc) and those from the forelimb located medially within VPLc. The densest labeling was found ventrally in VPLc and dorsally within ventroposterior inferior n. (VPI) only after injections of the forelimb. A more scattered, dorsal distribution of labeling was seen in the rest of VPLc from injections involving more proximal parts of the body representation in SII.     

Times of generation of glutamic acid decarboxylase immunoreactive neurons in mouse somatosensory cortex

The birth dates of neurons showing glutamic acid decarboxylase (GAD) immunoreactivity have been determined in mouse somatosensory cortex. Pregnant C57Bl mice received pulse injections of (3H)thymidine from E10 through E17 (E0 being the day of mating). The distributions of thymidine-labeled, GAD-positive and nonimmunoreactive (non-GAD) cells as a function of depth under the pial surface were recorded in adult animals. The maximum rate of generation of GAD-positive neurons occurred at E14, whereas the generation of non-GAD neurons reached its maximum rate at E13. Except for those in layer I, GAD-positive neurons followed an inside-out sequence of positioning. GAD-positive neurons born at E12 and E13 were located in layers VI-IV. GAD-positive neurons born at E14 were found throughout the cortical thickness, with a maximum in layer IV. The GAD-positive neurons labeled after pulses at E15 or E16 or E17 were limited to the superficial strata, forming a band that became narrower as it moved toward the pial surface with increase in age of pulse labeling. GAD-positive neurons in layer I were generated at a constant rate during the whole embryonic period analyzed. Non-GAD neurons also followed an inside-out spatiotemporal gradient. Two partially overlapping phases were distinguished in non-GAD neurogenesis. During the first phase (from E12 to E14) neurons populating adult layers VI and V originated, while neurons located in layers IV through I were generated during the second phase (from E13 to E17). Since GAD-immunoreactive neurons form a heterogeneous population, we envisage further studies in order to test whether differences exist in birth dates among the classes.     

Developmental history of the transient subplate zone in the visual and somatosensory cortex of the macaque monkey and human brain

The cytological organization and the timetable of emergence and dissolution of the transient subplate zone subjacent to the developing visual and somatosensory cortex were studied in a series of human and monkey fetal brains. Cerebral walls processed with Nissl, Golgi, electron-microscopic, and histochemical methods show that this zone consists of migratory and postmigratory neurons, growth cones, loosely arranged axons, dendrites, synapses, and glial cells. In both species the subplate zone becomes visible at the beginning of the mid-third of gestation as a cell-poor/fiber-rich layer situated between the intermediate zone and the developing cortical plate. The subplate zone appears earlier in the somatosensory than in the visual area and reaches maximal width at the beginning of the last third of gestation in both regions. At the peak of its size the ratio between the width of the subplate zone and cortical plate in the somatosensory cortex is 2:1 in monkey and 4:1 in man while in the occipital lobe these structures have about equal width in both species. The dissolution of the subplate zone begins during the last third of gestation with degeneration of some subplate neurons and the relocation of fiber terminals into the cortex. The subplate zone disappears faster in the visual than in the somatosensory area. The present results together with our previous findings support the hypothesis that the subplate zone may serve as a "waiting" compartment for transient cellular interactions and a substrate for competition, segregation, and growth of afferents originated sequentially from the brain stem, basal forebrain, thalamus, and from the ipsi- and contralateral cerebral hemisphere. After a variable and partially overlapping time period, these fibers enter the cortical plate while the subplate zone disappears leaving only a vestige of cells scattered throughout the subcortical white matter. A comparison between species indicates that the size and duration of the subplate zone increases during mammalian evolution and culminates in human fetuses concomitantly with an enlargement of cortico-cortical fiber systems. The regional difference in the size, pattern, and resolution of the subplate zone correlates also with the pattern of cerebral convolutions. Our findings indicate that, contrary to prevailing notions, the subplate may not be a vestige of the phylogenetically old network but a transient embryonic structure that expanded during evolution to subserve the increasing number of its connections.