Areas PEc Research Articles

Similar neural states, but dissimilar decoding patterns for motor control in parietal cortex.

Discrete neural states are associated with reaching movements across the fronto-parietal network. Here, the Hidden Markov Model (HMM) applied to spiking activity of the somato-motor parietal area PE revealed a sequence of states similar to those of the contiguous visuomotor areas PEc and V6A. Using a coupled clustering and decoding approach, we proved that these neural states carried spatiotemporal information regarding behaviour in all three posterior parietal areas. However, comparing decoding accuracy, PE was less informative than V6A and PEc. In addition, V6A outperformed PEc in target inference, indicating functional differences among the parietal areas. To check the consistency of these differences, we used both a supervised and an unsupervised variant of the HMM, and compared its performance with two more common classifiers, Support Vector Machine and Long-Short Term Memory. The differences in decoding between areas were invariant to the algorithm used, still showing the dissimilarities found with HMM, thus indicating that these dissimilarities are intrinsic in the information encoded by parietal neurons. These results highlight that, when decoding from the parietal cortex, for example, in brain machine interface implementations, attention should be paid in selecting the most suitable source of neural signals, given the great heterogeneity of this cortical sector.

Decoding sensorimotor information from superior parietal lobule of macaque via Convolutional Neural Networks

Despite the well-recognized role of the posterior parietal cortex (PPC) in processing sensory information to guide action, the differential encoding properties of this dynamic processing, as operated by different PPC brain areas, are scarcely known. Within the monkey’s PPC, the superior parietal lobule hosts areas V6A, PEc, and PE included in the dorso-medial visual stream that is specialized in planning and guiding reaching movements. Here, a Convolutional Neural Network (CNN) approach is used to investigate how the information is processed in these areas. We trained two macaque monkeys to perform a delayed reaching task towards 9 positions (distributed on 3 different depth and direction levels) in the 3D peripersonal space. The activity of single cells was recorded from V6A, PEc, PE and fed to convolutional neural networks that were designed and trained to exploit the temporal structure of neuronal activation patterns, to decode the target positions reached by the monkey. Bayesian Optimization was used to define the main CNN hyper-parameters. In addition to discrete positions in space, we used the same network architecture to decode plausible reaching trajectories. We found that data from the most caudal V6A and PEc areas outperformed PE area in the spatial position decoding. In all areas, decoding accuracies started to increase at the time the target to reach was instructed to the monkey, and reached a plateau at movement onset. The results support a dynamic encoding of the different phases and properties of the reaching movement differentially distributed over a network of interconnected areas. This study highlights the usefulness of neurons’ firing rate decoding via CNNs to improve our understanding of how sensorimotor information is encoded in PPC to perform reaching movements. The obtained results may have implications in the perspective of novel neuroprosthetic devices based on the decoding of these rich signals for faithfully carrying out patient’s intentions.

The Superior Parietal Lobule of Macaque Monkey: Relative Influence of Gaze and Static Arm Position during Reaching.

The superior parietal lobule (SPL) integrates somatosensory, motor, and visual signals to dynamically control arm movements. During reaching, visual and gaze signals are used to guide the hand to the desired target location, while proprioceptive signals allow to correct arm trajectory, and keep the limb in the final position at the end of the movement. Three SPL areas are particularly involved in this process: V6A, PEc, PE. Here, we evaluated the influence of eye and arm position on single neuron activity of these areas during the holding period at the end of arm reaching movements, when the arm is motionless and gaze and hand positions are aligned. Two male macaques (Macaca fascicularis) performed a foveal reaching task while single unit activity was recorded from areas V6A, PEc, and PE. We found that at the end of reaching movements the neurons of all these areas were modulated by both eye position and static position of the arm. V6A and PEc showed a prevalent combination of gaze and proprioceptive input, while PE seemed to encode these signals more independently. Our results demonstrate that all these SPL areas combine gaze and proprioceptive input to provide an accurate monitoring of arm movements.

Real and Imagined Grasping Movements Differently Activate the Human Dorsomedial Parietal Cortex

Object prehension typically includes a transport phase (reaching) and a grip phase (grasping). Within the posterior parietal cortex (PPC), grasping movements have been traditionally associated to a lateral activation, although recent monkey evidence suggests also a medial involvement. Here, we wanted to determine whether grasping-related activities are present in the human dorsomedial parietal cortex, by focusing on two cortical regions specialized in the monkey in controlling limb movements, i.e., V6A (composed by its ventral and dorsal sectors, V6Av and V6Ad, respectively) and PEc, both recently defined also in humans. We acquired functional magnetic resonance images while participants performed both real (pantomimed) and imagined grasping of visually-presented objects. We found that the human areas V6Ad (hV6Ad) and PEc (hPEc) were both activated by real grasping, whereas hV6Ad only was activated by the imagery of grasping movements. hV6Av was not involved in either types of grasping. These results speak against the traditional notion of a medial-to-lateral segregation of reaching versus grasping information within the PPC and strengthen the idea that the human dorsomedial parietal cortex implements the whole complex pattern of visuomotor transformations required for object-oriented actions. Our findings suggest that hV6Ad is particularly involved in implementing all the visuomotor transformations needed to create an abstract representation of the object-directed action, while hPEc is involved in implementing the sensorimotor transformations needed to actually perform that action.

Receptor density pattern confirms and enhances the anatomic-functional features of the macaque superior parietal lobule areas

The macaque monkey superior parietal lobule (SPL) is part of a neuronal network involved in the integration of information from visual and somatosensory cortical areas for execution of reaching and grasping movements. We applied quantitative in vitro receptor autoradiography to analyse the distribution patterns of 15 different receptors for glutamate, GABA, acetylcholine, serotonin, dopamine, and adenosine in the SPL of three adult male Macaca fascicularis monkeys. For each area, mean (averaged over all cortical layers) receptor densities were visualized as a receptor fingerprint of that area. Multivariate analyses were conducted to detect clusters of areas according to the degree of (dis)similarity of their receptor organization. Differences in regional and laminar receptor distributions confirm the location and extent of areas V6, V6Av, V6Ad, PEc, PEci, and PGm as found in cytoarchitectonic and functional studies, but also enable the definition of three subdivisions within area PE. Receptor densities are higher in supra- than in infragranular layers, with the exception of kainate, M2, and adenosine receptors. Glutamate and GABAergic receptors are the most expressed in all areas analysed. Hierarchical cluster analyses demonstrate that SPL areas are organized in two groups, an organization that corresponds to the visual or sensory-motor characteristics of those areas. Finally, based on present results and in the framework of our current understanding of the structural and functional organization of the primate SPL, we propose a novel pattern of homologies between human and macaque SPL areas.

The neglected medial part of macaque area PE: segregated processing of reach depth and direction

Area PE (Brodmann's area 5), located in the posterior parietal cortex (PPC), is involved in the control of arm movements. Many monkey studies showed PE's involvement in reach directions, while only a few revealed signals coding the depth of reaches. Notably, all these studies focused on the lateral part of PE, leaving its medial part functionally largely unexplored. We here recorded neuronal activity in the medial part of PE in three male Macaca fascicularis while they performed coordinated eye and arm movements in darkness towards targets located at different directions and depths. We used the same task as in our previous studies of more caudal PPC sectors (areas V6A and PEc), allowing a direct comparison between these three PPC areas. We found that, in medial PE, reach direction and depth were encoded mainly by distinct populations of neurons. Directional signals were more prominent before movement onset, whereas depth processing occurred mainly during and after movement execution. Visual and somatosensory mapping of medial PE revealed a lack of visual responses yet strong somatosensory sensitivity, with a representation of both upper and lower limbs, distinct from the somatotopy reported in lateral PE. This study shows that PE is strongly involved in motor processing of depth and direction information during reaching. It highlights a trend in medial PPC, going from the joint coding of depth and direction signals caudally, in area V6A, to a largely segregated processing of the two signals rostrally, in area PE.

Thalamo-cortical projections to the macaque superior parietal lobule areas PEc and PE.

The exposed surface of the superior parietal lobule in macaque brain contains two architectonically defined areas named PEc and PE. The aim of the present study is the characterization of thalamic afferents of these two areas. For this purpose, retrograde neuronal tracers were injected, or placed in crystal form, in areas PEc and PE. We found that the two areas show a similar pattern of thalamic inputs, mainly originating from Lateral Posterior (LP), Pulvinar (Pul), Ventral Posterior Lateral (VPL), and Ventral Lateral (VL) nuclei, all structures known to be involved in visual, somatosensory, and/or sensorimotor processing. Minor afferents were observed from the Centromedian/Parafascicular complex (CM/PF), Central Lateral (CL), Ventral Anterior (VA), and Medial Dorsal (MD) nuclei. LP and VL were more strongly connected to PEc than to PE, while the other main thalamic inputs to the two areas showed slight differences in strength. The part of the Pul mostly connected with areas PEc and PE was the Medial Pul. No labeled cells were found in the retinotopically organized Lateral and Inferior Pul. In the somatotopically organized VPL and VL nuclei, labeled neurons were mainly found in regions likely to correspond to the trunk and limb representations (in particular the legs). These findings are in line with the sensory-motor nature of areas PEc and PE, and with their putative functional roles, being them suggested to be involved in the preparation and control of limb interaction with the environment, and in locomotion.

Sensory properties of the caudal aspect of the macaque's superior parietal lobule.

In the superior parietal lobule (SPL), the anterior part (area PE) is known to process somatosensory information, while the caudalmost part (areas V6Av and V6) processes visual information. Here we studied the visual and somatosensory properties of the areas PEc and V6Ad located in between the somatosensory and visual domains of SPL. About 1500 neurons were extracellularly recorded in 19 hemispheres of 12 monkeys (Macaca fascicularis). Visual and somatosensory properties of single neurons were generally studied separately, while in a subpopulation of neurons, both the sensory properties were tested. Visual neurons were more represented in V6Ad and somatosensory neurons in PEc. The visual neurons of these two areas showed similar properties and represented a large part of the contralateral visual field, mostly the lower part. In contrast, somatosensory neurons showed remarkable differences. The arms were overrepresented in both the areas, but V6Ad represented only the upper limbs, whereas PEc both the upper and lower limbs. Interestingly, we found that in both the areas, bimodal visual-somatosensory cells represented the proximal part of the arms. We suggest that PEc is involved in locomotion and in the control of hand/foot interaction with the objects of the environment, while V6Ad is in the control of the object prehension specifically performed with the upper limbs. Neuroimaging and lesion studies from literature support a strict homology with humans.

Claustral afferents of superior parietal areas PEc and PE in the macaque.

The exposed surface of the primate superior parietal cortex includes two cytoarchitectonically defined areas, the PEc and PE. In the present study we describe the distribution of neurons projecting from the claustrum to these areas. Retrograde neuronal tracers were injected by direct visualization of regions of interest, and the location of injection sites was reconstructed relative to cytoarchitectural borders. For comparison, the patterns of claustral label that resulted from injections involving neighboring cytoarchitectonic areas were analyzed. We found that the claustral territories sending projections to areas PE and PEc partially overlapped zones previously shown to form projections to the posterior parietal, somatosensory, visual, and motor cortex. The projection zones to the PE and PEc overlapped extensively, and consisted of multiple patches separated by label-free zones. Most of the labeled neurons were located in the posterior-ventral part of the claustrum. Area PE received additional inputs from a posterior-dorsal part of the claustrum, which has been previously reported to project to the somatosensory cortex, while the PEc receives additional input from an anterior-ventral region of the claustrum, which has been reported to project to the visual association cortex. These observations reflect the known functional properties of the PE and PEc, with the former containing neurons that are predominantly involved in somatosensory processing, and the latter including both somatosensory and visual neurons. The present results suggest that the claustrum projections may help coordinate the activity of an extensive neural circuit involved in sensory and motor processing for movement execution. J. Comp. Neurol. 525:1475-1488, 2017. © 2016 Wiley Periodicals, Inc.

Cortical connections of the visuomotor parietooccipital area V6Ad of the macaque monkey

Area V6A, a functionally defined region in the anterior bank of the parietooccipital sulcus, has been subdivided into dorsal and ventral cytoarchitectonic fields (V6Ad and V6Av). The aim of this study was to define the cortical connections of the cytoarchitectonic field V6Ad. Retrograde and bidirectional neuronal tracers were injected into the dorsal part of the anterior bank of parietooccipital sulcus of seven macaque monkeys (Macaca fascicularis). The limits of injection sites were compared with those of cytoarchitectonic fields. The major connections of V6Ad were with areas of the superior parietal lobule. The densest labeling was observed in the medial intraparietal area (MIP). Areas PEc, PGm, and V6Av were also strongly connected. Labeled cells were found in medial parietal area 31; in cingulate area 23; in the anterior (AIP), ventral (VIP), and lateral (LIP) intraparietal areas; in the inferior parietal lobule (fields Opt and PG); and in the medial superior temporal area (MST). In the frontal lobe, the main projection originated from F2, although labeled cells were also found in F7 and area 46. Preliminary results obtained from injections in nearby areas PEc and V6Av revealed connections different from those of V6Ad. In agreement with functional data, the strong connections with areas where arm-reaching activity is represented suggest that V6Ad is part of a parietofrontal circuit involved in the control of prehension, and connections with AIP specifically support an involvement in the control of grasping. Connections with areas LIP and Opt are likely related to the oculomotor activities observed in V6Ad.

Cortical afferents to the smooth-pursuit region of the macaque monkey’s frontal eye field

In primates, the frontal eye field (FEF) contains separate representations of saccadic and smooth-pursuit eye movements. The smooth-pursuit region (FEFsem) in macaque monkeys lies principally in the fundus and deep posterior wall of the arcuate sulcus, between the FEF saccade region (FEFsac) in the anterior wall and somatomotor areas on the posterior wall and convexity. In this study, cortical afferents to FEFsem were mapped by injecting retrograde tracers (WGA-HRP and fast blue) into electrophysiologically identified FEFsem sites in two monkeys. In the frontal lobe, labeled neurons were found mostly on the ipsilateral side in the (1) supplementary eye field region and lateral area F7; (2) area F2 along the superior limb of the arcuate sulcus; and (3) in the buried cortex of the arcuate sulcus extending along the superior and inferior limbs and including FEFsac and adjacent areas 8, 45, and PMv. Labeled cells were also found in the caudal periprincipal cortex (area 46) in one monkey. Labeled cells were found bilaterally in the frontal lobe in the deep posterior walls of the arcuate sulcus and postarcuate spurs and in cingulate motor areas 24 and 24c. In postcentral cortical areas all labeling was ipsilateral and there were two major foci of labeled cells: (1) the depths of the intraparietal sulcus including areas VIP, LIP, and PEa, and (2) the anterior wall and fundus of the superior temporal sulcus including areas PP and MST. Smaller numbers of labeled cells were found in superior temporal sulcal areas FST, MT, and STP, posterior cingulate area 23b, area 3a within the central sulcus, areas SII, RI, Tpt in the lateral sulcus, and parietal areas 7a, 7b, PEc, MIP, DP, and V3A. Many of these posterior afferent cortical areas code visual-motion (MT, MST, and FST) or visual-motion and vestibular (PP, VIP) signals, consistent with the responses of neurons in FEFsem and with the overall physiology and anatomy of the smooth-pursuit eye movement system.

空間識・シリーズ教育講座 １１．頭頂野の入出力構造

Recently, many functional areas have been identified in the parietal cortex of the brain in monkeys, and owing to this, cytoarchitectonic subdivision has been elaborated in the monkey parietal cortex. This review deals with sulcal patterns of the parietal lobe in human and monkey brains, and corticocortical connections of both neurophysiologi-cally defined regions and cytoarchitectonic areas of the parietal cortex in monkeys. The author proposes that the transverse occipital sulcus is a boundary between the parietal and occipital cortices in the human brain. A brain map of the monkey is presented. The parietal cortical areas are subdivided into somatosensory, eye movement related, hand or upper limb movement related, vestibular, and auditory areas, according to neu-rophysiologically defined functions of individual cortical areas. The somatosensory system contains areas 3a, 3b, 1, 2, 5, 7b, and SII: S1 projects to motor and premotor areas and somatosensory association areas project to premotor areas, the prefrontal, temporal, and limbic cortices, and the hippocampus. The eye movement system is composed of areas 7a, LIP, and PIP: It receives input from visual and temporal cortices and sends output to premotor areas, the prefrontal cortex, and the hippocampus. Hand movements are mediated in areas AIP, VIP, CIP, MIP, PEc, and V6A: These areas receive afferents from somatosensory areas, visual and temporal cortices and send efferents to the ventral premotor area. The vestibular functions are executed by coordination of areas 3aNV, 3aHV, 2NV, Ri, and VPS: These areas reciprocate with the cingulate cortex and insula and project to the ventral premotor area, SWA, and frontal eye field. The auditory system involves areas VIP and LIP: It receives afferents from the superior temporal sulcal cortex and sends efferents to the premotor area, frontal eye field, and prefrontal cortex.

Eye-hand coordination during reaching. II. An analysis of the relationships between visuomanual signals in parietal cortex and parieto-frontal association projections.

The relationships between the distribution of visuomanual signals in parietal cortex and that of parieto-frontal projections are the subject of the present study. Single cell recording was performed in areas PEc and V6A, where different anatomical tracers were also injected. The monkeys performed a variety of behavioral tasks, aimed at studying the visual and motor properties of parietal cells, as well as the potential combination of retinal-, eye- and hand-related signals on cell activity. The activity of most cells was related to the direction of movement and the active position of the hand. Many of these reach-related cells were influenced by eye position information. Fewer cells displayed relationships to saccadic eye movements. The activity of most neurons related to a combination of both hand and eye signals. Many cells were also modulated during preparation for hand movement. Light-dark differences of activity were common and interpreted as related to the sight and monitoring of hand motion and/or position in the visual field. Most cells studied were very sensitive to moving visual stimuli and also responded to optic flow stimulation. Visual receptive fields were generally large and extended to the periphery of the visual field. For most neurons, the orientation of the preferred directions computed across different epochs and tasks conditions clustered within a limited sector of space, the field of global tuning. This can be regarded as an ideal frame to combine spatially congruent eye- and hand-related information for different forms of visuomanual behavior. All these properties were common to both PEc and V6A. Retinal, eye- and hand-related activity types, as well as parieto-frontal association cells, were distributed in a periodic fashion across the tangential domain of areas PEc and V6A. These functional and anatomical distributions were characterized and compared through a spectral and coherency analysis, which revealed the existence of a selective 'match' between activity types and parieto-frontal connections. This match depended on where each individual efferent projection was addressed. The results of the present and of the companion study can be relevant for a re-interpretation of optic ataxia as the consequence of the breakdown of the combination of retinal-, eye- and hand-related directional signals within the global tuning fields of parietal neurons.

Eye-hand coordination during reaching. I. Anatomical relationships between parietal and frontal cortex.

The anatomical and physiological substrata of eye-hand coordination during reaching were studied through combined anatomical and physiological techniques. The association connections of parietal areas V6A and PEc, and those of dorso-rostral (F7) and dorso-caudal (F2) premotor cortex were studied in monkeys, after physiological characterization of the parietal regions where retrograde tracers were injected. The results show that parieto-occipital area V6A is reciprocally connected with F7, and receives a smaller projection from F2. Local parietal projections to V6A arise from areas MIP and, to a lesser extent, 7m, PEa and PEC: On the contrary, parietal area PEc is strongly and reciprocally connected with the part of F2 located close to the pre-central dimple (pre-CD). Local parietal projections to PEc come from a distributed network, including PEa, MIP, PEci and, to a lesser extent, 7m, V6A, 7a and MST. Premotor area F7 receives parietal projections mainly from 7m and V6A, and local frontal projections mainly from F2. On the contrary, premotor area F2 in the pre-CD zone receives parietal inputs from PEc and, to a lesser extent, PEci, while in the peri-arcuate zone F2 receives parietal projections from PEa and MIP. Local frontal projections to F2 pre-CD mostly stem from F4, and, to a lesser extent, from F7 and F3, and CMAd; those addressed to peri-arcuate zone of F2 arise mainly from F5 and, to a lesser extent, from F7, F4, dorsal (CMAd) and ventral (CMAv) cingulate motor areas, pre-supplementary (F6) and supplementary (F3) motor areas. The distribution of association cells in both frontal and parietal cortex was characterized through a spectral analysis that revealed an arrangement of these cells in the form of bands, composed of cell clusters, or 'columns'. The reciprocal connections linking parietal and frontal cortex might explain the presence of visually related and eye-position signals in premotor cortex, as well as the influence of information about arm position and movement direction in V6A and PEC: The association connections identified in this study might carry sensory as well motor information that presumably provides a basis for a re-entrant signaling. This might be necessary to match retinal-, eye- and hand-related information underlying eye-hand coordination during reaching.

Superior area 6 afferents from the superior parietal lobule in the macaque monkey

Superior area 6 of the macaque monkey frontal cortex is formed by two cytoarchitectonic areas: F2 and F7. In the present experiment, we studied the input from the superior parietal lobule (SPL) to these areas by injecting retrograde neural tracers into restricted parts of F2 and F7. Additional injections of retrograde tracers were made into the spinal cord to define the origin of corticospinal projections from the SPL. The results are as follows: 1) The part of F2 located around the superior precentral dimple (F2 dimple region) receives its main input from areas PEc and PEip (PE intraparietal, the rostral part of area PEa of Pandya and Seltzer, [1982] J. Comp. Neurol. 204:196-210). Area PEip was defined as that part of area PEa that is the source of corticospinal projections. 2) The ventrorostral part of F2 is the target of strong projections from the medial intraparietal area (area MIP) and from the dorsal part of the anterior wall of the parietooccipital sulcus (area V6A). 3) The ventral and caudal parts of F7 receive their main parietal input from the cytoarchitectonic area PGm of the SPL and from the posterior cingulate cortex. 4) The dorsorostral part of F7, which is also known as the supplementary eye field, is not a target of the SPL, but it receives mostly afferents from the inferior parietal lobule and from the temporal cortex. It is concluded that at least three separate parietofrontal circuits link the superior parietal lobule with the superior area 6. Considering the functional properties of the areas that form these circuits, it is proposed that the PEc/PEip-F2 dimple region circuit is involved in controlling movements on the basis of somatosensory information, which is the traditional role proposed for the whole dorsal premotor cortex. The two remaining circuits appear to be involved in different aspects of visuomotor transformations.

Projections to the frontal cortex from the posterior parietal region in the rhesus monkey

The projections to the frontal cortex from the various subdivisions of the posterior parietal region in the rhesus monkey were studied by means of autoradiographic technique. The rostral superior parietal lobule (area PE) projects to the dorsal areas 4 and 6 on the lateral surface of the frontal lobe as well as to the supplementary motor area (MII) on its medial surface. The caudal area PE sends its connections to dorsal area 6 and MII. The projections from the medial parietal cortex (areas PEc and PGm) are similar to those of the superior parietal lobule but they tend to concentrate in the more rostral part of dorsal area 6, MII, and in the cingulate gyrus (area 24). The most caudal part of the medial parietal cortex also projects to area 8. The anteriormost part of the inferior parietal lobule (area PF) projects to the ventral area 6, including the caudal bank of the lower branch of the arcuate sulcus, to the ventral area 46 below the sulcus principalis, and to the frontal and pericentral opercular cortex. The middle inferior parietal lobule (areas PFG and PG) projects to the ventral part of area 46 and area 8, whilst the posteriormost inferior parietal lobule (caudal PG and area Opt) is connected with both dorsal and ventral area 46, dorsal area 8, as well as the anteriormost dorsal area 6, and the cingulate gyrus (area 24).

Intrinsic connections and architectonics of posterior parietal cortex in the rhesus monkey

By means of autoradiographic and ablation-degeneration techniques, the intrinsic cortical connections of the posterior parietal cortex in the rhesus monkey were traced and correlated with a reappraisal of cerebral architectonics. Two major rostral-to-caudal connectional sequences exist. One begins in the dorsal postcentral gyrus (area 2) and proceeds, through architectonic divisions of the superior parietal lobule (areas PE and PEc), to a cortical region on the medial surface of the parietal lobe (area PGm). This area has architectonic features similar to those of the caudal inferior parietal lobule (area PG). The second sequence begins in the ventral post/central gyrus (area 2) and passes through the rostral inferior parietal lobule (areas PG and PFG) to reach the caudal inferior parietal lobule (area PG). Both the superior parietal lobule and the rostral inferior parietal lobule also send projections to various other zones located in the parietal opercular region, the intraparietal sulcus, and the caudalmost portion of the cingulate sulcus. Areas PGm and PG, on the other hand, project to each other, to the cingulate region, to the caudalmost portion of the superior temporal gyrus, and to the upper bank of the superior temporal sulcus. Finally, a reciprocal sequence of connections, directed from caudal to rostral, links together many of the above-mentioned parietal zones. With regard to the laminar pattern of termination, the rostral-to-caudal connections are primarily distributed in the form of cortical "columns" while the caudal-to-rostral connections are found mainly over the first cortical cell layer.