The distribution of corticocortical, thalamocortical, and callosal inputs on identified motor cortex output neurons: mechanisms for their selective recruitment.

Motor cortex neurons were identified antidromically in anesthetized cats by their axonal projections to one of six targets: (1) somatosensory cortex, (2) opposite motor cortex, (3) red nucleus, (4) lateral reticular nucleus, (5) spinal cord, and (6) ventrolateral thalamus. Three inputs to motor cortex were tested for their influences on the identified cortical efferent neurons. The tested inputs originated from ipsilateral somatosensory cortex, opposite motor cortex, and ventral thalamus. Subthreshold effects of input pathways were detected by monitoring latency variations of antidromic responses. The three afferent sources, when activated by electrical stimulation, were not equally effective on motor cortex neurons. Ipsilateral corticocortical and thalamocortical excitation were found for the majority of neurons; the influenced proportions ranged from 55% to 100%, according to the target of the output neurons. Effects from the opposite hemisphere were found for only 5% to 30% of the neurons in the same projection classes. Many neurons (36 of 81, or 44%) were excited from more than one source, but few (5 of 37, or 14%) were influenced by all three possible sources of input, even in small regions of cortex innervated by all three of the inputs. Among 19 electrode tracks where all three inputs were present, there were only 2 tracks where all the neurons shared the same combination of inputs. Even for neurons in closest anatomical proximity ("clusters"), it was unusual (only 7 of 25 clusters) for all the neurons to have the same input pattern. Among the seven clusters where all the neurons shared the same input pattern, five of the clusters projected to the same target. These variable combinations of inputs to motor cortex neurons support the conclusion that efferent neurons could be recruited selectively from separate cortical layers or from within clusters of nearby neurons, according to the target of their axonal projection.     

Experience with the use of horseradish peroxidase for light and electron microscope studies of interneuronal connections]

By the method based on a retrograde axonal transport of exogenous horseradish peroxidase (HRP), the origins of afferentation of the motor cortex of adult cats, kittens and albino rats were studied. HRP-positive neurons were found by light and electron microscopy in the somatosensory cortex (C1) of the ipsilateral hemisphere and in the portions of the cortex of the contralateral hemisphere which were symmetrical to the site of injection of HRP. The disposition of neurons, marked by HRP, in the Vth layer of the motor cortex suggest that these neurons may send their axons into the bundles of comissural fibres going to the motor cortex of the opposite hemisphere. This method considerably expands possibilities of revealing the origins of afferentation of the investigated portion of the nervous system and allows more complete and reliable investigation of interneuronal connections.     

Intrinsic properties of and thalamocortical inputs onto identified corticothalamic-VPM neurons

Corticothalamic (CT) feedback plays an important role in regulating the sensory information that the cortex receives. Within the somatosensory cortex layer VI originates the feedback to the ventral posterior medial (VPM) nucleus of the thalamus, which in turn receives sensory information from the contralateral whiskers. We examined the physiology and morphology of CT neurons in rat somatosensory cortex, focusing on the physiological characteristics of the monosynaptic inputs that they receive from the thalamus. To identify CT neurons, rhodamine microspheres were injected into VPM and allowed to retrogradely transport to the soma of CT neurons. Thalamocortical slices were prepared at least 3 days post injection. Whole-cell recordings from labeled CT cells in layer VI demonstrated that they are regular spiking neurons and exhibit little spike frequency adaption. Two anatomical classes were identified based on their apical dendrites that either terminated by layer V (compact cells) or layer IV (elaborate cells). Thalamic inputs onto identified CT-VPM neurons demonstrated paired pulse depression over a wide frequency range (2–20?Hz). Stimulus trains also resulted in significant synaptic depression above 10?Hz. Our results suggest that thalamic inputs differentially impact CT-VPM neurons in layer VI. This characteristic may allow them to differentiate a wide range of stimulation frequencies which in turn further tune the feedback signals to the thalamus.     

Attention-related,cross-modality modulation of somatosensory neurons in primate ventrobasal (VB) thalamus

Attention-related modulation (AM) of the somatosensory responses of single neurons has been demonstrated in the cerebral cortex and medullary dorsal horn, but not in the ventrobasal thalamus. The somatically evoked activity was recorded of single units in the ventral posterior lateral thalamus (VPL) of awake monkeys while they detected the termination of task-relevant somatic or visual stimuli. Eighteen of 56 somatically responsive VPL neurons are reported that were recorded for enough time for a complete analysis of their responses during both the visual and somatic attention tasks. All neurons were spontaneously active and responded either to innocuous cutaneous (13/18) or deep (5/18) stimuli. Seven neurons (7/18, 38.8%) showed AM of somatosensory responsiveness. Two cells (2/7, 28.6%) showed AM only during the visual task, two others (2/7, 28.6%) only during the somatosensory task, and three cells (3/7, 42.8%) showed AM during both tasks. All five cells showing AM during the somatosensory task had enhanced responses to the task-relevant somatic stimulus. In contrast, the somatosensory responses of all five cells showing AM during the visual task were reduced. It is concluded that selective attention is associated with a modality specific modulation of the somatosensory responses of a sub-population of neurons within the primate VPL nucleus.     

Horseradish peroxidase study of connections of the hippocampal (mediodorsal) cortex ofOphisaurus apodus

Sources of afferent projections of the hippocampal (mediodorsal) cortex were detected in lizards (Ophisaurus apodus) by the retrograde horseradish peroxidase transport method. Labeled neurons after injection of the enzyme were most numerous in the anterior dorsolateral thalamic nucleus, mammillary body, superior nucleus raphe, dorsal cortex (ipsilaterally), and the hippocampal cortex of the contralateral hemisphere. Fewer neurons projecting into the hippocampal cortex were found in these same structures on the side opposite to that of the injection, and also in the ventromedial zone of the telencephalon (olfactory tubercle, the nucleus of Broca's diagonal band, and the nucleus accumbens), the preoptic region of the hypothalamus, and the ventrotegmental region of the midbrain. Endings of efferent fibers from the hippocampal cortex were found in the septum, thalamus, hypothalamus (mainly on the side of injection of the enzyme), and also in the hippocampal and dorsal cortex of both hemispheres. The results show that afferent and efferent connections of the lizard's hippocampal cortex are similar to those of mammals.     

Attention-related, cross-modality modulation of somatosensory neurons in primate ventrobasal (VB) thalamus

Attention-related modulation (AM) of the somatosensory responses of single neurons has been demonstrated in the cerebral cortex and medullary dorsal horn, but not in the ventrobasal thalamus. The somatically evoked activity was recorded of single units in the ventral posterior lateral thalamus (VPL) of awake monkeys while they detected the termination of task-relevant somatic or visual stimuli. Eighteen of 56 somatically responsive VPL neurons are reported that were recorded for enough time for a complete analysis of their responses during both the visual and somatic attention tasks. All neurons were spontaneously active and responded either to innocuous cutaneous (13/18) or deep (5/18) stimuli. Seven neurons (7/18, 38.8%) showed AM of somatosensory responsiveness. Two cells (2/7, 28.6%) showed AM only during the visual task, two others (2/7, 28.6%) only during the somatosensory task, and three cells (3/7, 42.8%) showed AM during both tasks. All five cells showing AM during the somatosensory task had enhanced responses to the task-relevant somatic stimulus. In contrast, the somatosensory responses of all five cells showing AM during the visual task were reduced. It is concluded that selective attention is associated with a modality specific modulation of the somatosensory responses of a sub-population of neurons within the primate VPL nucleus.     

Morphological and physiological study of afferent projections of the posterolateral thalamic nucleus in rats

It was shown by the method of retrograde axonal transport of horseradish peroxidase that the posterolateral thalamic nucleus (NPL) in rats receives considerable ascending projections from the superior colliculus (SC), the dorsal part of the lateral geniculate body (LGB), and the pretectal region (PT) and smaller projections from n. ventralis posterior (VP) and n. ventralis lateralis (VL) of the thalamus, the ventral part of LGB, the zona incerta, and anterior hypothalamus. The most marked descending projections run into NPL from area 18A of the cortex and the dentate fascia of the hippocampus, whereas inputs from cortical areas 18, 20, 7, 29c, 17, and 36 are less marked. In electrophysiological experiments with peripheral stimulation of visual, auditory, and somatosensory systems, polysensory convergence and interaction between signals from these systems were studied during isolated and simultaneous presentation of heterosensory stimuli. Of 229 neurons tested, 134 (58.5%) responded to at least one of the stimuli mentioned. Among monomodal neurons (53 of 134) there were some cells which responded to visual (77.4%) and somatic (22.6%) stimulation; neurons which responded only to acoustic stimulation were not found in the nucleus. As far as polymodal neurons (81 of 134) responding to two or three sensory stimuli are concerned, the most effective inputs of these units were visual and somatosensory. Interaction between stimuli acting on polymodal neurons was expressed as mutual inhibition or facilitation of responses; opposite effects could be observed on the various components of these responses.I. M. Sechenov Institute of Evolutionary Physiology and Biochemistry, Academy of Sciences of the USSR, Leningrad. Translated from Neirofiziologiya, Vol. 16, No. 2, pp. 168–176, March–April, 1984.     

Convergence across Tactile Afferent Types in Primary and Secondary Somatosensory Cortices

Integration of information by convergence of inputs onto sensory cortical neurons is a requisite for processing higher-order stimulus features. Convergence across defined peripheral input classes has generally been thought to occur at levels beyond the primary sensory cortex, however recent work has shown that this does not hold for the convergence of slowly-adapting and rapidly-adapting inputs in primary somatosensory cortex. We have used a new analysis method for multi-unit recordings, to show convergence of inputs deriving from the rapidly-adapting and Pacinian channels in a proportion of neurons in both primary and secondary somatosensory cortex in the anaesthetised cat. We have validated this method using single-unit recordings. The secondary somatosensory cortex has a greater proportion of sites that show convergence of this type than primary somatosensory cortex. These findings support the hypothesis that the more complex features processed in higher cortical areas require a greater degree of convergence across input classes, but also shows that this convergence is apparent in the primary somatosensory cortex.     

In vivo and in vitro patch-clamp recording analysis of the process of sensory transmission in the spinal cord and sensory cortex

Following the integration and modification of the sensory inputs in the spinal cord, the information is transmitted to the primary sensory cortex where the integrated information is further processed and perceived. Processing of the sensory information in the spinal cord has been intensively investigated. However, the mechanisms of how the inputs are processed in the cortex are still unclear. To know the correlation of the sensory processing in the dorsal horn and cortex, in vivo and in vitro patch-clamp recordings were made from rat dorsal horn and sensory cortex. Although dorsal horn neurons showed spontaneous and evoked EPSCs by noxious and non-noxious stimuli, most somatosensory neurons located at 100 to 1000 microm from the surface of the cortex exhibited an oscillatory activity and received synaptic inputs from non-noxious but not noxious receptors. These observations suggest that the synaptic responses in cortical neurons are processed in a more complex manner; and this may be due to the reciprocal synaptic connection between thalamus and cortex.     

Organization of the cortico-ponto-cerebellar pathway to the dorsal paraflocculus. An experimental study with anterograde and retrograde transport of WGA-HRP in the cat.

To reveal the organization and relative magnitude of connections from various parts of the cerebral cortex to the dorsal paraflocculus via the pontine nuclei, WGA-HRP was injected in the dorsal paraflocculus in conjunction with injection of the same tracer in various parts of the cerebral cortex in 17 cats. Termination areas of cortical fibres (anterogradely labelled) and pontine neurons projecting to the dorsal paraflocculus (retrogradely labelled) were carefully plotted in serial transverse sections. As an average of countings in ten cats, 90% of the labelled cells were found in the pontine nuclei contralateral to the injection, and the majority (70%) were located in the rostral half of the nuclei. The highest degree of overlap between anterograde and retrograde labelling was found after injections of the parietal association cortex (areas 5 and 7). In an experiment with double anterograde tracing, it was shown that both area 5 and 7 contribute substantially to the cerebral inputs to the dorsal paraflocculus. High degree of overlap also occurred after injections of several visual cortical areas (areas 17, 18, 19, 20 and the posteromedial lateral suprasylvian visual area, PMLS). Cases with injections restricted to individual visual areas indicate that they all contribute to the parafloccular input. Considerably less overlap occurred after injections of the primary sensorimotor region (SI, MI) and second somatosensory area (SII), while the supplementary motor area, the auditory cortex and gyrus cinguli probably have no or very restricted access to the dorsal paraflocculus. It is concluded that the dorsal paraflocculus has its major cortical input from the parietal association cortex and the visual cortical areas. Since all the various cortical regions studied project to largely different parts of the pontine nuclei, and overlap with neurons projecting to the dorsal paraflocculus takes place at numerous places, it follows that the pontine neurons projecting to the dorsal paraflocculus must consist of many subgroups differing with regard to their cortical input.     

Whole-brain mapping of direct inputs to midbrain dopamine neurons

Recent studies indicate that dopamine neurons in the ventral tegmental area (VTA) and substantia nigra pars compacta (SNc) convey distinct signals. To explore this difference, we comprehensively identified each area's monosynaptic inputs using the rabies virus. We show that dopamine neurons in both areas integrate inputs from a more diverse collection of areas than previously thought, including autonomic, motor, and somatosensory areas. SNc and VTA dopamine neurons receive contrasting excitatory inputs: the former from the somatosensory/motor cortex and subthalamic nucleus, which may explain their short-latency responses to salient events; and the latter from the lateral hypothalamus, which may explain their involvement in value coding. We demonstrate that neurons in the striatum that project directly to dopamine neurons form patches in both the dorsal and ventral striatum, whereas those projecting to GABAergic neurons are distributed in the matrix compartment. Neuron-type-specific connectivity lays a foundation for studying how dopamine neurons compute outputs.     

Some introductory notes on the organization of the forebrain in tetrapods.

As an introduction to the main theme of this conference an overview of the organization of the tetrapod forebrain is presented with emphasis on the telencephalic representation of sensory and motor functions. In all classes of tetrapods, olfactory, visual, octavolateral, somatosensory and gustatory information reaches the telencephalon. Major differences exist in the telencephalic targets of sensory information between amphibians and amniotes. In amphibians, three targets are found: the lateral pallium for olfactory input, the medial pallium for visual and multisensory input, and the lateral subpallium for visual, octavolateral and somatosensory information. The forebrains of reptiles and mammals are similar in that the dorsal surface of their cerebral hemisphere is formed by a pallium with three major segments: (a) an olfactory, lateral cortex; (b) a 'limbic' cortex that forms the dorsomedial wall of the hemisphere, and (c) an intermediate cortex that is composed entirely of isocortex in mammals, but in reptiles (and birds) consists of at least part of the dorsal cortex (in birds the Wulst) and a large intraventricular protrusion, i.e. the dorsal ventricular ridge. In birds, the entire lateral wall of the hemisphere is involved in this expansion. The intermediate pallial segment receives sensory projections from the thalamus and contains modality-specific sensory areas in reptiles, birds and mammals. The most important differences between the intermediate pallial segment of amniotes concern motor systems.     

Multiple Representations of the Body and Input-Output Relationships in the Agranular and Granular Cortex of the Chronic Awake Guinea Pig

The organization of somatosensory input and the input-output relationships in regions of the agranular frontal cortex (AGr) and granular parietal cortex (Gr) were examined in the chronic awake guinea pig, using the combined technique of single-unit recording and intracortical microstimulation (ICMS). AGr, which was cytoarchitectonically subdivided into medial (AGrm) and lateral (AGrl) parts, also can be characterized on a functional basis. AGrl contains the head, forelimb, and most hindlimb representations; only a small number of hindlimb neurons are confined in AGrm. Different distributions of submodalities exist in AGr and Gr: AGr receives predominantly deep input (with the exception of the vibrissa region, which receives cutaneous input), whereas neurons of Gr respond almost exclusively to cutaneous input. The cutaneous or deep receptive field (RF) of each neuron was determined by natural peripheral stimulation. All studied neurons were activated by small RFs, with the exception of lip, nose, pinna, and limb units of lateral Gr (Grl), for which the RFs were larger.

Microelectrode mapping experiments revealed the existence of three spatially separate, incomplete body maps in which somatosensory and motor representations overlap. One body map, with limbs medially and head rostrolaterally, is contained in AGr. A second map, comparable to the first somatosensory cortex (SI) of other mammals, is found in Gr, with hindlimb, trunk, forelimb, and head representations in an orderly mediolateral sequence. An unresponsive zone separates the head area from the forelimb region. A third map, with the forelimb rostrally and the hindlimb caudally, lies adjacent and lateral to the SI head area. This limb representation, which is characterized by an upright and small size compared to that found in SI, can be considered to be part of the second somatosensory cortex (SII). A distinct head representation was not recognized as properly belonging to SII, but the evidence that neurons of the SI head region respond to stimulation of large RFs located in lips, nose, and pinna leads us to hypothesize that the SII face area overlaps that of SI to some extent, or, alternatively, that the two areas are strictly contiguous and the limits are ambiguous, making them difficult to distinguish.

The input-output relationships were based on the results of RF mapping and ICMS in the same electrode penetration. The intrinsic specific interconnections of cortical neurons whose afferent input and motor output is related to identical body regions show a considerable degree of refinement. The input-output correspondence is especially pronounced for neurons with small RFs. This study confirms and extends similar data recently reported for other rodents.     

Axonal conduction properties of antidromically identified neurons in rat barrel cortex

Physiological studies of the rodent somatosensory cortex have consistently described considerable heterogeneity in receptive field properties of neurons outside of layer IV, particularly those in layers V and VI. One such approach for distinguishing among different local circuits in these layers may be to identify the projection target of neurons whose axon collaterals contribute to the local network. In vivo, this can be accomplished using antidromic stimulation methods. Using this approach, the axonal conduction properties of cortical efferent neurons are described. Four projection sites were activated using electrical stimulation: (1) vibrissal motor cortex, (2) ventrobasal thalamus (VB), (3) posteromedial thalamic nucleus (POm), and (4) cerebral peduncle. Extracellular recordings were obtained from a total of 169 units in 21 animals. Results demonstrate a close correspondence between the laminar location of the antidromically identified neurons and their anatomically known layer of origin. Axonal properties were most distinct for corticofugal axons projecting through the crus cerebri. Corticothalamic axons projecting to either VB or POm were more similar to each other in terms of laminar location and conduction properties, but could be distinguished using focal electrical stimulation. It is concluded that, once stimulation parameters are adjusted for the small volume of the rat brain, the use of antidromic techniques may be an effective strategy to differentiate among projection neurons comprising different local circuits in supra- and infragranular circuits.     

Functional properties of efferent zones of the motor cortex which project to ipsilateral and contralateral face muscles

In both intact (4 animals) and lesioned (2 preparations with contralateral motor cortex ablation and 1 animal with transection of the rostral two thirds of the corpus callosum) cats, three different types of efferent zones were identified in the face motor cortex by the technique of microstimulation: contralateral, ipsilateral and bilateral efferent zones. The three types of efferent zones had different organizational features such as location, thresholds of effective sites and latencies of motor responses. Mean thresholds of effective sites from ipsilateral and bilateral efferent zones in lesioned animals were not significantly higher than those in intact preparations. In both intact and lesioned animals, neurons endowed with contralateral, bilateral and ipsilateral receptive fields were isolated from the three types of efferent zones.     

Axonal conduction properties of antidromically identified neurons in rat barrel cortex

Physiological studies of the rodent somatosensory cortex have consistently described considerable heterogeneity in receptive field properties of neurons outside of layer IV, particularly those in layers V and VI. One such approach for distinguishing among different local circuits in these layers may be to identify the projection target of neurons whose axon collaterals contribute to the local network. In vivo, this can be accomplished using antidromic stimulation methods. Using this approach, the axonal conduction properties of cortical efferent neurons are described. Four projection sites were activated using electrical stimulation: (1) vibrissal motor cortex, (2) ventrobasal thalamus (VB), (3) posteromedial thalamic nucleus (POm), and (4) cerebral peduncle. Extracellular recordings were obtained from a total of 169 units in 21 animals. Results demonstrate a close correspondence between the laminar location of the antidromically identified neurons and their anatomically known layer of origin. Axonal properties were most distinct for corticofugal axons projecting through the crus cerebri. Corticothalamic axons projecting to either VB or POm were more similar to each other in terms of laminar location and conduction properties, but could be distinguished using focal electrical stimulation. It is concluded that, once stimulation parameters are adjusted for the small volume of the rat brain, the use of antidromic techniques may be an effective strategy to differentiate among projection neurons comprising different local circuits in supra- and infragranular circuits.     

The effects of DBS patterns on basal ganglia activity and thalamic relay

Thalamic neurons receive inputs from cortex and their responses are modulated by the basal ganglia (BG). This modulation is necessary to properly relay cortical inputs back to cortex and downstream to the brain stem when movements are planned. In Parkinson's disease (PD), the BG input to thalamus becomes pathological and relay of motor-related cortical inputs is compromised, thereby impairing movements. However, high frequency (HF) deep brain stimulation (DBS) may be used to restore relay reliability, thereby restoring movements in PD patients. Although therapeutic, HF stimulation consumes significant power forcing surgical battery replacements, and may cause adverse side effects. Here, we used a biophysical-based model of the BG-Thalamus motor loop in both healthy and PD conditions to assess whether low frequency stimulation can suppress pathological activity in PD and enable the thalamus to reliably relay movement-related cortical inputs. We administered periodic pulse train DBS waveforms to the sub-thalamic nucleus (STN) with frequencies ranging from 0-140 Hz, and computed statistics that quantified pathological bursting, oscillations, and synchronization in the BG as well as thalamic relay of cortical inputs. We found that none of the frequencies suppressed all pathological activity in BG, though the HF waveforms recovered thalamic reliability. Our rigorous study, however, led us to a novel DBS strategy involving low frequency multi-input phase-shifted DBS, which successfully suppressed pathological symptoms in all BG nuclei and enabled reliable thalamic relay. The neural restoration remained robust to changes in the model parameters characterizing early to late PD stages.     

Coping with variability in small neuronal networks

Experimental and corresponding modeling studies indicate that there is a 2- to 5-fold variation of intrinsic and synaptic parameters across animals while functional output is maintained. Here, we review experiments, using the heartbeat central pattern generator (CPG) in medicinal leeches, which explore the consequences of animal-to-animal variation in synaptic strength for coordinated motor output. We focus on a set of segmental heart motor neurons that all receive inhibitory synaptic input from the same four premotor interneurons. These four premotor inputs fire in a phase progression and the motor neurons also fire in a phase progression because of differences in synaptic strength profiles of the four inputs among segments. Our work tested the hypothesis that functional output is maintained in the face of animal-to-animal variation in the absolute strength of connections because relative strengths of the four inputs onto particular motor neurons is maintained across animals. Our experiments showed that relative strength is not strictly maintained across animals even as functional output is maintained, and animal-to-animal variations in strength of particular inputs do not correlate strongly with output phase. Further experiments measured the precise temporal pattern of the premotor inputs, the segmental synaptic strength profiles of their connections onto motor neurons, and the temporal pattern (phase progression) of those motor neurons all in the same animal for a series of 12 animals. The analysis of input and output in this sample of 12 individuals suggests that the number (four) of inputs to each motor neuron and the variability of the temporal pattern of input from the CPG across individuals weaken the influence of the strength of individual inputs. Moreover, the temporal pattern of the output varies as much across individuals as that of the input. Essentially, each animal arrives at a unique solution for how the network produces functional output.     

Dynamics of changes in somatosensory input to the motor cortex following lesion of somatosensory cortical areas in dogs

The pattern of change produced in somatosensory evoked potential (EP) in the forelimb projection area within the motor cortex (MI) following lesion of the projection area of the same limb in the somatosensory cortex (SI) or in parietal cortex area 5 was investigated during chronic experiments on waking dogs. Amplitude of the initial positive — negative wave of EP declined to 28–63% of preoperational level in all cases. No significant recovery of EP was noted for three weeks. Thus, a correlation between change in EP and spontaneous recuperation of the precision motor response occurring within two weeks after lesion of the SI did not exist. Nor was EP reinstated in the MI after ablation of area 5, despite complete but gradual reinstatement of EP (after an initial decline to 53%) in the nearby SI region. This protracted depression of EP seems to have been associated with breakdown of somatotopic sensory input from the SI or from area 5 to the MI, since EP in the motor cortex of the intact hemisphere and the hindlimb projection area within the MI on the lesioned side either remained unchanged or recovered within a week or two.Institute of Higher Nervous Activity and Neurophysiology, Academy of Sciences of the USSR, Moscow. Translated from Neirofiziologiya, Vol. 22, No. 1, pp. 61–68, January–February, 1990.     

A Comparison of the Ipsilateral Cortical Projections to the Dorsal and Ventral Subdivisions of the Macaque Premotor Cortex

The cortical connections of the dorsal (PMd) and ventral (PMv) subdivisions of the premotor area (PM, lateral area 6) were studied in four monkeys (Macaca fascicularis) through the use of retrograde tracers. In two animals, tracer was injected ventral to the arcuate sulcus (PMv), in a region from which forelimb movements could be elicited by intracortical microstimulation (ICMS). Tracer injections dorsal to the arcuate sulcus (PMd) were made in two locations. In one animal, tracer was injected caudal to the genu of the arcuate sulcus (in caudal PMd [cPMd], where ICMS was effective in eliciting forelimb movements); in another animal, it was injected rostral to the genu of the arcuate sulcus (in rostral PMd [rPMd], where ICMS was ineffective in eliciting movements). Retrogradely labeled neurons were counted in the ipsilateral hemisphere and located in cytoarchitectonically identified areas of the frontal and parietal lobes. Although both PMv and PMd were found to receive inputs from other motor areas, the prefrontal cortex, and the parietal cortex, there were differences in the topography and the relative strength of projections from these areas.

There were few common inputs to PMv and PMd; only the supplementary eye fields projected to all three areas studied. Interconnections within PMd or PMv appeared to link hindlimb and forelimb representations, and forelimb and face representations; however, connections between PMd and PMv were sparse. Areas cPMd and PMv were found to receive inputs from other motor areas—the primary motor area, the supplementary motor area, and the cingulate motor area—but the topography and strength of projections from these areas varied. Area rPMd was found to receive sparse inputs, if any, from these motor areas. The frontal eye field (area 8a) was found to project to PMv and rPMd, and area 46 was labeled substantially only from rPMd. Parietal projections to PMv were found to originate from a variety of somatosensory and visual areas, including the second somatosensory cortex and related areas in the parietal operculum of the lateral sulcus, as well as areas 5, 7a, and 7b, and the anterior intraparietal area. By contrast, projections to cPMd arose only from area 5. Visual areas 7m and the medial intraparietal area were labeled from rPMd. Relatively more parietal neurons were labeled after tracer injections in PMv than in PMd. Thus, PMv and PMd appear to be parts of separate, parallel networks for movement control.