==== Front Proc Natl Acad Sci U S AProc. Natl. Acad. Sci. U.S.ApnaspnasPNASProceedings of the National Academy of Sciences of the United States of America0027-84241091-6490National Academy of Sciences 2999717420180392610.1073/pnas.1803926115Biological SciencesNeuroscienceSocial SciencesPsychological and Cognitive SciencesPlasticity based on compensatory effector use in the association but not primary sensorimotor cortex of people born without hands http://orcid.org/0000-0002-5900-3455Striem-Amit Ella a1Vannuscorps Gilles abcCaramazza Alfonso adaDepartment of Psychology, Harvard University, Cambridge, MA 02138;bFaculty of Psychology and Educational Sciences, Psychological Sciences Research Institute Université Catholique de Louvain, Louvain-la-Neuve 1348, Belgium;cInstitute of Neurosciences, Université Catholique de Louvain, Louvain-la-Neuve 1348, Belgium;dCenter for Mind/Brain Sciences, University of Trento, 38068 Rovereto, Italy1To whom correspondence should be addressed. Email: striemamit@fas.harvard.edu.Edited by Jon H. Kaas, Vanderbilt University, Nashville, TN, and approved June 20, 2018 (received for review March 5, 2018) Author contributions: E.S.-A., G.V., and A.C. designed research; E.S.-A. and G.V. performed research; E.S.-A. analyzed data; and E.S.-A., G.V., and A.C. wrote the paper. 24 7 2018 11 7 2018 11 7 2018 115 30 7801 7806 Copyright © 2018 the Author(s). Published by PNAS.2018This open access article is distributed under Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND).Significance What determines the role of brain regions and their plasticity when typical inputs or experience is not provided? To what extent can extreme compensatory use affect brain organization? We tested the reorganization of primary and association sensorimotor cortex hand-selective areas in people born without hands, who use their feet for everyday tasks. We found that their primary sensorimotor hand area is preferentially activated for nearby body parts that cannot serve as effectors. In contrast, foot-selective compensatory plasticity was found in the association cortex, in an area typically involved in manual tool use. This shows limitations of compensatory plasticity and experience in modifying brain organization of early topographical cortex, as compared with association cortices where function-based organization is the driving factor. What forces direct brain organization and its plasticity? When brain regions are deprived of their input, which regions reorganize based on compensation for the disability and experience, and which regions show topographically constrained plasticity? People born without hands activate their primary sensorimotor hand region while moving body parts used to compensate for this disability (e.g., their feet). This was taken to suggest a neural organization based on functions, such as performing manual-like dexterous actions, rather than on body parts, in primary sensorimotor cortex. We tested the selectivity for the compensatory body parts in the primary and association sensorimotor cortex of people born without hands (dysplasic individuals). Despite clear compensatory foot use, the primary sensorimotor hand area in the dysplasic subjects showed preference for adjacent body parts that are not compensatorily used as effectors. This suggests that function-based organization, proposed for congenital blindness and deafness, does not apply to the primary sensorimotor cortex deprivation in dysplasia. These findings stress the roles of neuroanatomical constraints like topographical proximity and connectivity in determining the functional development of primary cortex even in extreme, congenital deprivation. In contrast, increased and selective foot movement preference was found in dysplasics’ association cortex in the inferior parietal lobule. This suggests that the typical motor selectivity of this region for manual actions may correspond to high-level action representations that are effector-invariant. These findings reveal limitations to compensatory plasticity and experience in modifying brain organization of early topographical cortex compared with association cortices driven by function-based organization. brain developmenthandsmotor cortexplasticitysomatosensory cortexEC | H2020 | H2020 Priority Excellent Science | H2020 Marie Skłodowska-Curie Actions (MSCA)100010665654837Ella Striem-Amit ==== Body What determines the role of brain regions and their plasticity when typical inputs or experience is not provided? To what extent can extreme compensatory use affect brain organization? Is brain organization limited in its plasticity due to strong neuroanatomical constraints? We investigated this question by studying the neural activity profile in primary and association sensorimotor cortices in people born without hands who use their feet to perform everyday manual tasks. Both animal studies and human cases of deafferentation or amputation have found that the primary sensory and motor cortices reorganize to show activity expansion of adjacent body parts, and that this reorganization is greater when it occurs early in life (1–10). However, several recent functional neuroimaging studies in people with congenitally missing hands (dysplasic individuals) have stressed the role of experience-based plasticity in the reorganization of the primary sensorimotor hand region. Specifically, it has been proposed that the hand area supports the use of other body parts in performing manual-type, everyday actions, and that this plasticity is not constrained by topographical factors. Results have been reported showing activation of the hand primary sensorimotor cortex for body parts used in place of congenitally missing hands (11, 12). The hand-responsive area in somatosensory cortex of individuals with partially missing hands shrinks depending on the size and use of the hand remains (13, 14). Importantly, in dysplasic individuals, somatosensory stimulation of the foot activated the lateral sensory cortex (15). Furthermore, in two individuals born without hands bilaterally, it has been shown that strong transcranial magnetic stimulation of the lateral motor cortex generated motor evoked potentials not only in the participants’ residual finger or shoulder but also in the foot. Such stimulation also interfered with performance of a foot motor task (11). These findings were interpreted as evidence for robust plasticity of the sensorimotor cortex and functional takeover of the hand areas by the foot. These studies were interpreted as raising the possibility that the primary sensorimotor cortex is functionally selective rather than selective for a specific body part. On this view, the hand area stands for any effector that functions as a hand in everyday tasks like grasping and manipulating objects (12). However, the specificity of such supposed compensatory reorganization (for the body part used as hands) has not been thoroughly tested in people born without hands. None of the studies with dysplasic individuals tested whether the hand area is activated more for compensatorily used body parts than for other proximal, but noncompensatorily used organs. Is the hand sensorimotor cortex indeed selective for compensatorily used body parts that serve as effectors? Alternatively, plasticity due to compensatory effector experience may be limited by neuroanatomical constraints such as topographical proximity and connectivity of this brain system, enabling only a takeover by closer cortical territories or contralateral intact body parts, akin to what is found in late-onset amputation or deafferentation. A related question concerns the effects of missing limbs on the organization of the association sensorimotor action system. Parts of the posterior parietal cortex show preference for specific body parts, specifically the eyes and hands, when participating in goal-directed visuomotor action (16–20). These preferences in the fronto-parietal network also depend on the functional use of body parts such as in the execution of manual tasks like reaching, object manipulation, and grasping (21–25). Given that the foot rather than the hand is the major effector in dysplasics, we additionally tested whether typically hand-selective regions in the parietal and frontal cortex now respond preferentially to the foot, reflecting the effects of effector use-based compensatory plasticity. We investigated these questions by mapping sensorimotor responses to movement of various body parts in five individuals born without hands who use their feet for everyday functions. The results are discussed in the context of the broader issue of the roles of neural proximity and functional equivalence in shaping the reorganization of primary and association cortices in the absence of relevant sensorimotor experience. Results To test the specificity of plasticity in the sensorimotor hand areas in people born without arms or hands (SI Appendix, Table S1), we used an active motor paradigm, designed to activate both primary somatosensory and primary motor cortices, similar to previous studies of reorganization (refs. 12 and 26–29; for supporting findings in a passive somatosensory stimulation design, see below). We scanned five dysplasic subjects, as well as a control group, as they performed simple flexing movements of different body parts. The body parts chosen for this experiment included the hands (in the controls, used as a localizer) and the feet, which the dysplasic subjects use to overcome their disability. Our study participants, according to self-report, rely largely on their dexterous feet (dominantly their right foot; all were right-footed) to perform daily typical manual activities. Their feet are extraordinarily dexterous, allowing them to use cell phones, utensils and nearly all other everyday tools (see list at SI Appendix, Table S2). In a questionnaire of tool use, all of the dysplasic subjects reported using their lower limbs for most of the tools they have used (SI Appendix, Fig. S1). Foot tool use accounted for a minimum of 92% of the used tools, although few tools were jointly manipulated by the lower face or remaining upper limbs in specific individuals (SI Appendix, Fig. S1 and Table S1). Additionally, we inspected movement of the shoulder and lips, expected to activate neighboring cortical regions on both sides of the missing hand territory, as well as a control body part, which is not an immediate cortical neighbor of the hand and is not used compensatorily to replace hand function as a dexterous effector: the abdomen. While abdominal core muscles may be used to stabilize the body and may thus be used in excess by the dysplasics while using their feet, they cannot be used to replace hand function. In the control group, this protocol resulted in a typical somatotopic activation pattern with sensorimotor responses along a superior-inferior axis for the foot, abdomen, shoulder, hand, and lips movements (Fig. 1A; hand peak is delineated in white), replicating the known Penfield homunculus (30). Topographic responses were found along the central sulcus, postcentral gyrus, and postcentral sulcus, covering Brodmann areas 3a, 3b, and 1 (31), and extending anteriorly from the central sulcus also to the precentral gyrus, covering areas 4a and 4p (32) and potentially parts of the premotor cortex. In the dysplasic subjects, although the peak responses remained topographic along the primary sensorimotor cortex, movement in every tested body part extended toward the deprived area and generated some activation in the hand region (Fig. 1B). This also included increased activation of the primary hand area while they moved their right foot, used by these subjects to perform typically manual actions, as reported before (11, 12). Indeed, plotting the group differences showed that the sensorimotor hand area had stronger activation in the dysplasic subjects than in the controls for moving the various body parts, including the foot (Fig. 1 C and D). Fig. 1. Activation for multiple body parts in the sensorimotor hand area in dysplasic subjects (born without hands). (A) The activation for body part movement (contraction of the lips, right shoulder, abdomen, and right foot) in the typically developed control subject group (random effect GLM analysis; P < 0.05 corrected for multiple comparisons) is shown on the left cortical hemisphere, following the standard Penfield homunculus. The sensorimotor hand area, delineated in white, represents the core area activated by right hand movement in all of the control participants (each at P < 0.05 corrected). This area encompasses both primary somatosensory and motor areas and extends beyond them to the postcentral sulcus posteriorly and precentral gyrus anteriorly. CS, central sulcus; IPS, intraparietal sulcus; PoCS, postcentral sulcus. (B) Activation for body part movement is shown for the dysplasic individuals, born without hands (fixed effect GLM analysis, P < 0.05 corrected; see SI Appendix, Table S1 for subject upper limb structure). Movement of each of the tested body parts elicited activation in the hand-selective area to some extent, including, as previously reported, movement of the right foot. (C and D) Group difference in activation of body part movement is shown in both Bayesian analysis (C; BF10 > 10 represents strong evidence for the existence of a group difference) and frequentist statistics (D; mixed effects F test). The sensorimotor hand area is activated more by the dysplasic group for multiple body parts. Group differences extend somewhat more inferiorly in the posterior somatosensory cortex (postcentral gyrus) compared with the motor cortex, in accordance with a larger cortical representation to the upper limbs in this area (82). When exploring selectivity of sensorimotor responses, plotting the preferential activation per cortical vertex (in a winner-takes-all approach; Fig. 2 A and B), the dysplasic subjects show a preference for shoulder and abdomen movements in the typical hand area (Fig. 2B; hand area delineated white), which seem to have been displaced and expanded toward the hand area compared with the activation pattern in the controls. This reorganization for the shoulder appears to be similar across the sensory and motor cortices, although a preference for the abdomen is found more anteriorly, in the central sulcus itself, likely originating from the motor cortex. These patterns were consistent across the individual subjects (SI Appendix, Fig. S2) and found also in the right hemisphere for movement of the contralateral body parts (see SI Appendix, Fig. S3 for comparable analyses of the right hemisphere). We further sampled the response pattern within the hand region of interest (ROI). This ROI was defined as significant overlapping activation for hand movement, compared with baseline, in all control subjects, covering both primary sensory and motor cortex. The activation in the dysplasics’ primary hand region is strongest for the movement of the shoulder, which is proximal to the missing hand (and therefore also activates the hand region in controls; see also Figs. 1A and 2C), but is also significant for the abdomen. Crucially, moving the foot does not generate strong significant activation in the hand area at large (Fig. 2C). Despite the preferential compensatory use of the right foot in these subjects, movements of the foot do not seem to favorably overtake this region. In fact, movement of the shoulder and abdomen activated the hand region in the dysplasics significantly more than foot movement (P < 0.005 for both). Similarly, in a whole brain analysis, when contrasting moving the foot compared with the noncompensatory but more proximal abdomen, the dysplasic subjects do not show significant activation in their primary sensorimotor hand area, and activity was found for the reverse contrast (Fig. 2D; see SI Appendix, Fig. S5 A and C for unsmoothed individual subjects data and overlap analysis). Fig. 2. Selectivity for the compensatorily used foot in the dysplasics is found in associative somatosensory cortex, but not primary sensorimotor hand area. (A) Preferred body part responses for contraction movements (winner-takes-all approach) for the control subjects follows the standard Penfield homunculus. The sensorimotor hand area is delineated in white, representing the core area activated by right hand movement in all of the control participants (each at P < 0.05 corrected), as in Fig. 1. For a presentation on grooved (uninflated) cortical reconstruction, see SI Appendix, Fig. S6B, and for an overlay of probabilistic cytoarchitectonic parcellation, see SI Appendix, Fig. S7B. CS, central sulcus; IPS, intraparietal sulcus; PoCS, postcentral sulcus. (B) Preferred body part responses for contraction movements for the dysplasic group shows a preference for shoulder (and to some extent abdomen, in the motor cortex; see below for details) movements in the hand area, despite the extensive use of the feet to perform typically manual fine-motor tasks. Preferential activation for abdomen movement was found also on the anterior inferior border of the hand region, in the central sulcus, in agreement with evidence of a potential discontinuity in the motor cortex surrounding the hand area (83, 84). This abdomen preference is not found in passive tactile stimulation of the body; compare with SI Appendix, Fig. S4. Extensive preference for the foot is found outside the primary sensorimotor hand area, in the depth of the postcentral sulcus (potentially area BA 2) and extending to the aIPS. (C) Sensorimotor responses were sampled from the hand area, showing that this region in the dysplasics is more activated by proximal body parts (shoulder and abdomen/trunk) than by foot movements (P < 0.005 for both comparisons). Error bars for the control group (orange bars) represent SEM. Individual data points (blue diamonds) are presented for the five dysplasic individuals in addition to the group average. (D) Foot movement selectivity (over abdomen movement, representing a control body part that does not serve compensatorily as an effector) in the dysplasics can be found in the postcentral sulcus, intraparietal sulcus, superior parietal lobule, and premotor cortex (PMd), but not in the hand primary sensorimotor cortex, which shows the reverse preference. (E) Movement selectivity comparing the shoulder and foot in the dysplasics shows a robust preference to shoulder movement (a proximal, noncompensatory body part), rather than to foot movement, in the hand area. A preference for the foot is found in the aIPS. (F) Overall body part selectivity (comparing movement of all shared body parts; e.g., lips, shoulder, abdomen, and foot) differs between the dysplasics and controls (ANOVA body part x group interaction) in the inferior frontal lobe (including PMv) and in the sensorimotor hand area. (G) A direct comparison of the selectivity to right foot movement (vs. abdomen movement) between the dysplasics and control subjects shows potential for plasticity specific to the compensatorily used foot in the association cortices, in the inferior parietal lobule, in aIPS extending to the angular/supramarginal gyri, as well as the middle frontal gyrus, but not in the primary sensorimotor cortex. (H) Bayes factor (BF10) for difference between the groups in their differential activation to right foot movement (vs. abdomen movement) is shown. The dysplasics show different selectivity level for right foot movement compared with the controls in various cortical loci, including the sensorimotor hand area. However, the group difference found in the primary sensorimotor hand area in this analysis reflects a preference in the dysplasics group toward the abdomen movement (compare with D). Even more robustly, whole brain analysis contrasting foot movement to shoulder movement shows extensive significant selectivity for shoulder movement in the primary hand area of the dysplasics (Fig. 2E and SI Appendix, Fig. S5 B and D for individual subjects’ data and overlap analysis). Preference for the shoulder in the hand area was also found in the somatosensory cortex in a passive somatosensory stimulation experiment (see SI Appendix, Fig. S4 for right and left hemispheres; preference for the abdomen in the anterior motor hand area was not found for passive stimulation; compare with Fig. 2B). This does not mean that no foot selectivity exists in these subjects: When directly contrasting foot movement to movement of the abdomen in the dysplasic subjects (Fig. 2D), activation is found not only in the primary sensorimotor cortex foot region, which extends laterally from that of the controls, but also in the superior and inferior parietal lobule and posterior superior frontal sulcus. In the more stringent contrast (foot vs. shoulder), a significant cluster is found in the anterior intraparietal sulcus (aIPS), on its border with the postcentral sulcus. Overlaying our findings on a cytoarchitectonic parcellation probabilistic atlas (33–35) places this cluster on the aIPS at the border of Brodmann area 2 and area PFt in the inferior parietal lobule (see SI Appendix, Fig. S6 for findings overlaid on uninflated cortical reconstructions and SI Appendix, Fig. S7 for an overlay on the cytoarchitectonic parcellation probabilistic atlas). Body part selectivity differs between the dysplasics and controls also in the frontal lobe (including PMv) and in the sensorimotor hand area itself (ANOVA Body part X Group interaction; Fig. 2F). Both frequentist (Fig. 2G) and Bayesian statistics (Fig. 2H) of foot-selectivity group difference shows additional regions that may be related to compensatory use in the frontal (middle frontal gyrus) and parietal lobes. The inferior parietal cluster extends along the intraparietal sulcus (SI Appendix, Fig. S6), over the angular/supramarginal gyri and overlaps several cytoarchitectonic areas, including hIP1 and hIP2 (SI Appendix, Fig. S7). The more sensitive Bayesian analysis hints at group effects also in additional foci and in the temporal lobe, but these may require verification from additional studies, perhaps with increased group size. Therefore, plasticity related to extensive foot use in the dysplasics is found in the association parietal cortex, in the inferior parietal lobule (aIPS), and not in the primary sensorimotor hand area. We also inspected whether functional connectivity (FC) may reflect use-based plasticity within the primary sensorimotor cortex and in the association cortices. The group differences in FC from the hand area across the entire cortex showed FC differences only in the association cortex, in the middle intraparietal sulcus (mIPS; SI Appendix, Fig. S8). Discussion We tested people born without hands who use their feet to perform everyday manual tasks. We found that their primary sensorimotor hand area is activated by foot movements more than in typically developed control subjects (Fig. 1), replicating previous findings of plasticity in congenital limb absence (11, 12, 15). However, in contrast to previous research, we additionally tested for the selectivity of the compensatorily used effector and found that the hand area shows no foot selectivity. Instead, the hand area is significantly more activated by proximal but noncompensatory body parts, such as the shoulder and abdomen, which are not used as dexterous effectors (Fig. 2). In the association sensorimotor cortex, we found evidence for compensatory plasticity specific to the foot. Foot selectivity in the aIPS and inferior parietal lobule (Fig. 2 D, E, and G) was found exclusively in the dysplasic participants. Past findings of primary sensorimotor plasticity in people born without hands were thought to reflect use-dependent, compensatory plasticity for sensorimotor loss (11, 12, 15). Those results were interpreted as evidence that the primary sensorimotor cortex shows functional selectivity for performance of tasks typically conducted with the hands, and not necessarily for the specific body part (12). This interpretation builds on a model based on the organization of association sensory cortex in congenital blindness and deafness. In these cases, the role of association sensory cortices appears to be defined not strictly by their typical primary sensory input (e.g., vision) but by their function. This was demonstrated for high-level visual cortex in the blind where domain selectivity was found for distinct object domains, including artifacts, body parts, and scenes (36–42) with nonvisual stimuli. Similar results were obtained for nonvisual spatial localization (42) and motion perception (43). Analogous results of retained functional specialization independent of modality of input have been found for the deaf (44–46). These findings encouraged the view that the observation of seemingly compensatory reorganization in people born with partial dysplasia is driven by similarity of function as opposed to a topographically driven specialization in early sensorimotor cortex (12). Our findings of significant plasticity showing preference for body parts not used as effectors instead of compensatory foot selectivity, even in the complete absence of hands and in individuals who show strong compensatory strategies with foot use, point to important limitations of the function-specific model of primary sensorimotor cortex. The case of primary sensorimotor cortex in dysplasics differs in several respects from the findings in the blind and deaf. Beyond the different senses, these cases represent different stations of the cortical processing hierarchy. Specifically, in the blind and deaf, claims of functionally selective organization are limited to the associative sensory cortices, whereas plasticity in people born without hands was tested for the primary sensory-motor cortex. Therefore, while it may be that functionally driven plasticity principles could govern the reorganization of higher sensorimotor cortices, they do not seem to apply to the first cortical stations involved in processing touch and movement, which is governed by topographic mapping of the body. Consistent with these contrasting effects, we found foot selectivity in the dysplasics in the association parietal cortex, which may be indicative of compensatory plasticity and effector-function organization. An additional difference between these cases is the extent of the deprivation of the sensory-motor modality. In complete blindness or deafness, there are no competing inputs within the sensory modality in the early sensory cortices. Therefore, while the functional role of these regions is still debated (47–49), the topographic organization in the early stations of the hierarchy is retained (50–52). In higher stations of cortical processing, the visual and auditory cortices also receive inputs from other modalities and from downstream cortical stations via feedback connectivity, which become more dominant in the complete absence of visual input. These may then drive cortex organization toward similar functions and domains even in the absence of the typical visual/auditory features driving these regions. However, the case of absence of one body part leaves intact inputs from proximal body parts in the topographic organization, generating within-modality competition and overtake. In the somatosensory cortex, topographic overtake is mediated through horizontal, direct, and indirect (transsynaptic) connections (53) as well as from close body parts in subcortical nuclei. Moreover, in both motor and somatosensory cortex, there is overlap of the fields for different, close-by body parts (1, 53). This overlap is evident in our data: The hand area of control subjects is significantly activated as they move their shoulder (Fig. 2C), causing a limited activation difference between the groups for shoulder movement (Fig. 1C). The ability to evoke plasticity and takeover by nearby body parts has been confirmed in multiple studies of the effects of adult-stage and even early-life amputation or deafferentation (1–10, 54). Related findings of lesion-induced map plasticity and takeover in early sensory cortical areas following sensory loss in adulthood were also found for audition following cochlear lesions (resulting in tinnitus; ref. 55) and for vision following retinal lesions (56). Additionally, use-dependent plasticity in adults learning a new motor skill also allows for the expansion of relevant body parts (57, 58). It follows naturally that the same level of plasticity is possible in earlier development, and in the case of congenital limb absence, even without being strongly driven by extreme compensatory use. Importantly, topographical intramodal connectivity appears to more strongly drive the primary sensorimotor cortex than inputs from other functionally significant downstream stations or topographically remote body parts such as the foot, which could direct it toward compensatory roles. While some connectivity can be modulated to permit innervation from remote body parts in cases of deafferentation (6), our data suggest it is not sufficient, even in congenital bilateral upper limb dysplasia, to evoke preferential takeover. Thus, the current study reveals clear limitations for brain compensatory plasticity and for the role of experience in modifying brain organization, and highlights the topographical nature of the early sensory and motor cortices. In contrast to the results for primary sensorimotor cortex, we found increased selectivity for the foot in dysplasics’ association cortex. Specifically, we found selectivity for foot movement compared with movement of the abdomen and shoulder in parts of the inferior parietal lobule, mainly in the anterior IPS area, bordering the postcentral sulcus and, to some extent, in the angular/supramarginal gyri (Fig. 2 B, D, and E). We also found increased FC from the primary sensorimotor hand area to IPS, albeit in a more posterior region, in mIPS (SI Appendix, Fig. S8B). The strongest result was found in the aIPS and, inferiorly to it, in the inferior parietal lobule, falling with high probability within several cytoarchitectonic areas, from BA2 to hIP2 and PFt. Electrophysiological data in nonhuman primates and fMRI data in humans have shown that these regions support various types of manual behaviors. These include visually guided object manipulation, tool use and grasping movements (23, 25, 59–62), action observation and imitation (63), and additional more abstract action representations of function and goal (64, 65). Therefore, this region’s motor selectivity for the foot may be functional and related to compensatory use, linking it to a role in representing effector-invariant/effector-independent action; i.e., related to any body part used as an effector. Whereas several studies have shown effector-invariant action representations in the parietal lobe with regard to hand laterality and correspondence between hand and eye (66–69), generally in mIPS and pIPS, aIPS shows effector selectivity for the hand, compared with foot or eye movements (70). Importantly, aIPS appears to support grasping actions not only with the hands but also with tools (71–74). Our findings would therefore suggest that this region’s role may have to do with the use of the hands or tools as effectors, enabling it to also extend its selectivity to the foot, when it is used as an effector throughout life. Thus, the current study shows a clear division in organization principles between the primary, topographic cortices, and the association cortex. We find that these regions differ in their compensatory plasticity and in the manner in which experience can affect their organization. In the primary sensorimotor cortex, we find that similarly to the lesion-induced map plasticity in late-onset animal and human studies, even congenital deprivation and compensatory use cannot overcome neuroanatomical constraints such as topographical proximity and connectivity, which determine the functional development of these brain regions. In contrast, functional specialization occurs at relatively high levels of representation in the parietal lobe of the dysplasics. Analogously to the case of the association cortices of the blind and deaf, the aIPS in dysplasics appears to show effector-invariant plasticity and selectivity—to the foot in this case. This suggests its role extends beyond the hand to support object manipulation with other bodily effectors. Methods Participants. Five individuals born with severely shortened or completely absent upper limbs (individuals with upper limb dysplasia; dysplasics 1–5), and eight typically developed control subjects, matched for age (no group difference; P < 0.25) participated in the experiment. The causes of dysplasia were genetic, teratogenic medications (thalidomide), or unknown. See SI Appendix, Table S1 and Supplemental Methods for the summary of the characteristics of the dysplasics, as well as images of their residual limbs. None of the dysplasics had a history of phantom limb sensations or movements, and all were adept at performing everyday actions and tool use with their feet (see SI Appendix, Table S2). All of the dysplasic participants used their right foot dominantly. None of participants had a history of psychiatric or neurological disorder. They gave written informed consent in accordance with the institutional review board of Harvard University, which approved all of the experiments. Experimental Design. The motor experiment was carried out in a block design fMRI experiment. Mouth, abdomen and either side hands (for the control subjects), shoulders, and feet were moved (simple flexing/contraction movement) in separate blocks (6 s movement and 6 s rest) in randomized order according to an auditory cue (metronome). Four flex and relax movements were performed in each block at a frequency of 0.66 Hz. Due to our focus on the compensatory use of the feet, and as all of the dysplasic participants were dominantly right-footed, we used the movements of the right hand and foot for further examination and provide evidence for similar organization of the right hemisphere in response to left hand and foot movement in SI Appendix, Fig. S3. A supplementary somatosensory experiment with four of the five dysplasic subjects was carried out in a block design fMRI experiment (see acquisition detail and detail of the paradigm in SI Appendix, Supplemental Methods). Functional Imaging. The blood oxygen level-dependent (BOLD) fMRI measurements were obtained in a Siemens Trio 3-T scanner at the Center for Brain Science at Harvard University. For acquisition detail, see SI Appendix, Supplemental Methods. The main experiment had three runs of 186 whole-brain images each collected in one functional scan. Separate 3D recordings were used for coregistration and surface reconstruction. Data analysis was performed using the Brain Voyager QX 2.8 software package (Brain Innovation) using standard preprocessing procedures (SI Appendix, Supplemental Methods). Functional and anatomical datasets for each subject were aligned and fit to standardized Talairach space (75). Anatomical cortical reconstruction procedures included the segmentation of the white matter using a grow-region function embedded in Brain Voyager. The Talairach normalized cortical surface was then used for surface-based alignment conducted across the subjects according to their cortical curvature (sulci and gyri) patterns. All further analyses were conducted in cortical space. Single-subject data were spatially smoothed with a 2D 4 vertex full-width at half-maximum Gaussian to reduce intersubject anatomical variability for group analysis. Unsmoothed data are presented at the single-subjects cortical level (SI Appendix, Figs. S2 and S5). Due to the small sample size of the unique dysplasic group, analyses were based on converging evidence from the group analyses (general linear model; GLM; Figs. 1 A and B and 2 A, B, D, and E and SI Appendix, Fig. S3 A, B, D, and E), single-subject level (SI Appendix, Figs. S2 and S5), mixed effects ANOVA (Figs. 1D and 2 F and G and SI Appendix, Fig. S3 F and H), and Bayesian analysis (Figs. 1C and 2H and SI Appendix, Fig. S3G) to enable an assessment of the consistency of the findings. The minimum significance level of the results presented in this study (both individual and group level analyses) was set at P < 0.05, corrected for multiple comparisons, using the spatial extent method, a set-level statistical inference correction (76). The hand sensorimotor cortex (delineated in white in Figs. 1 and 2) was defined according to a full overlap (100%) of activation to right hand flexing across all control subjects (each at P < 0.05 corrected) to overcome potential individual biases. This region was further used to sample movement responses in each group (Fig. 2C) and as a seed to compute FC (SI Appendix, Fig. S8). Somatotopic preferential mapping was computed at a surface level for each dysplasic subject (SI Appendix, Fig. S2). Each cortical vertex is colored based on the body part whose movement elicited the highest activation (GLM estimate, beta value). Group analyses in the control group were conducted in a hierarchical random effects analysis (RFX GLM; ref. 77) and, in the dysplasics, a fixed-effect GLM was implemented, due to the group size. For group level somatotopic preferential mapping, each cortical vertex is colored based on the body part whose movement elicited the highest average activation across the group (Fig. 2 A and B). The same analyses were performed for the somatosensory supplemental experiment (SI Appendix, Fig. S4). Group analyses are presented on the Colin27 brain inflated (or folded; SI Appendix, Fig. S6) cortices, to which individual surface (cortical) data were aligned based on the curvature patterns. To link our findings to the anatomic characterization of these regions, activation for some of the contrasts was overlaid on the probabilistic cytoarchitectonic atlas (33–35, 78) (SI Appendix, Figs. S6 and S7). All areas defined by these maps are shown at 40% probability of belonging to this cytoarchitectonic region. Group comparisons were conducted using both frequentist (t test and mixed effects ANOVA; with group and body part factors; Figs. 1D and 2F and SI Appendix, Fig. S3F) and sensitive Bayesian analyses (refs. 79 and 80, Figs. 1C and 2H and SI Appendix, Figs. S3G and S8B), appropriate for testing small samples of unique populations and patients. See detail of the Bayesian analyses in SI Appendix, Supplemental Methods. Bayes factor (BF) of over 3 is considered substantial evidence and BF over 10 is considered strong evidence against the null hypothesis (79), in our case suggesting a group difference. The data generated and analyzed for the current study are available from the corresponding author upon request. FC. A dataset of spontaneous BOLD fluctuations for the investigation of intrinsic (rest state; ref. 81) FC was collected while the subjects lay in the scanner with no external stimulation or task. For details, see SI Appendix, Supplemental Methods. Supplementary Material Supplementary File We thank the dysplasic subjects who participated in our experiments, Himanshu Bhat and Thomas Benner of Siemens Healthcare for the simultaneous multislice echo-planar imaging (SMS-EPI) sequence, Steven Cauley of Massachusetts General Hospital for modifications that enabled implementation of our protocols in a routine session, and Tamar Makin for her helpful comments on an earlier version of the manuscript. This work was supported by Società Scienze Mente Cervello–Fondazione Cassa di Risparmio di Trento e Rovereto, a grant from the Provincia Autonoma di Trento, and a Harvard Provostial postdoctoral fund (to A.C.), and by the European Union’s Horizon 2020 Research and Innovation Programme under Marie Sklodowska-Curie Grant Agreement 654837 and the Israel National Postdoctoral Award Program for Advancing Women in Science (to E.S.-A.). The authors declare no conflict of interest. This article is a PNAS Direct Submission. This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1803926115/-/DCSupplemental. ==== Refs 1 Sanes JN Donoghue JP 2000 Plasticity and primary motor cortex Annu Rev Neurosci 23 393 415 10845069 2 Hall EJ Flament D Fraser C Lemon RN 1990 Non-invasive brain stimulation reveals reorganized cortical outputs in amputees Neurosci Lett 116 379 386 2243618 3 Pons TP 1991 Massive cortical reorganization after sensory deafferentation in adult macaques Science 252 1857 1860 1843843 4 Merzenich MM 1983 Topographic reorganization of somatosensory cortical areas 3b and 1 in adult monkeys following restricted deafferentation Neuroscience 8 33 55 6835522 5 Kaas JH Merzenich MM Killackey HP 1983 The reorganization of somatosensory cortex following peripheral nerve damage in adult and developing mammals Annu Rev Neurosci 6 325 356 6340591 6 Jain N Qi H-X Collins CE Kaas JH 2008 Large-scale reorganization in the somatosensory cortex and thalamus after sensory loss in macaque monkeys J Neurosci 28 11042 11060 18945912 7 Schieber MH Deuel RK 1997 Primary motor cortex reorganization in a long-term monkey amputee Somatosens Mot Res 14 157 167 9402646 8 Flor H 1995 Phantom-limb pain as a perceptual correlate of cortical reorganization following arm amputation Nature 375 482 484 7777055 9 Philip BA Frey SH 2014 Compensatory changes accompanying chronic forced use of the nondominant hand by unilateral amputees J Neurosci 34 3622 3631 24599461 10 Makin TR 2013 Deprivation-related and use-dependent plasticity go hand in hand eLife 2 e01273 24220510 11 Stoeckel MC Seitz RJ Buetefisch CM 2009 Congenitally altered motor experience alters somatotopic organization of human primary motor cortex Proc Natl Acad Sci USA 106 2395 2400 19164537 12 Hahamy A 2017 Representation of multiple body parts in the missing hand territory of congenital onehanders Curr Biol 27 1350 1355 28434861 13 Stoeckel MC Jörgens S Witte OW Seitz RJ 2005 Reduced somatosensory hand representation in thalidomide-induced dysmelia as revealed by fMRI Eur J Neurosci 21 556 562 15673455 14 Stoeckel MC Pollok B Witte OW Seitz RJ Schnitzler A 2005 Shrinkage of somatosensory hand area in subjects with upper extremity dysmelia revealed by magnetoencephalography J Neurophysiol 93 813 818 15469960 15 Stoeckel MC Pollok B Schnitzler A Witte OW Seitz RJ 2004 Use-dependent cortical plasticity in thalidomide-induced upper extremity dysplasia: Evidence from somaesthesia and neuroimaging Exp Brain Res 156 333 341 14745466 16 Crawford JD Henriques DY Medendorp WP 2011 Three-dimensional transformations for goal-directed action Annu Rev Neurosci 34 309 331 21456958 17 Fattori P Gamberini M Kutz DF Galletti C 2001 ‘Arm-reaching’ neurons in the parietal area V6A of the macaque monkey Eur J Neurosci 13 2309 2313 11454035 18 Ferraina S 1997 Combination of hand and gaze signals during reaching: Activity in parietal area 7 m of the monkey J Neurophysiol 77 1034 1038 9065868 19 Dickinson AR Calton JL Snyder LH 2003 Nonspatial saccade-specific activation in area LIP of monkey parietal cortex J Neurophysiol 90 2460 2464 12801893 20 Calton JL Dickinson AR Snyder LH 2002 Non-spatial, motor-specific activation in posterior parietal cortex Nat Neurosci 5 580 588 12021766 21 Kaas JH Stepniewska I Gharbawie O 2012 Cortical networks subserving upper limb movements in primates Eur J Phys Rehabil Med 48 299 306 22407009 22 Fogassi L Luppino G 2005 Motor functions of the parietal lobe Curr Opin Neurobiol 15 626 631 16271458 23 Grefkes C Fink GR 2005 The functional organization of the intraparietal sulcus in humans and monkeys J Anat 207 3 17 16011542 24 Kaas JH Gharbawie OA Stepniewska I 2011 The organization and evolution of dorsal stream multisensory motor pathways in primates Front Neuroanat 5 34 21716641 25 Vingerhoets G 2014 Contribution of the posterior parietal cortex in reaching, grasping, and using objects and tools Front Psychol 5 151 24634664 26 Makin TR Scholz J Henderson Slater D Johansen-Berg H Tracey I 2015 Reassessing cortical reorganization in the primary sensorimotor cortex following arm amputation Brain 138 2140 2146 26072517 27 Lotze M Flor H Grodd W Larbig W Birbaumer N 2001 Phantom movements and pain. An fMRI study in upper limb amputees Brain 124 2268 2277 11673327 28 MacIver K Lloyd DM Kelly S Roberts N Nurmikko T 2008 Phantom limb pain, cortical reorganization and the therapeutic effect of mental imagery Brain 131 2181 2191 18567624 29 Foell J Bekrater-Bodmann R Diers M Flor H 2014 Mirror therapy for phantom limb pain: Brain changes and the role of body representation Eur J Pain 18 729 739 24327313 30 Penfield W Boldrey E 1937 Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation Brain 60 389 443 31 Geyer S Schleicher A Zilles K Microstructural Organization and Interindividual Variability 1999 Areas 3a, 3b, and 1 of human primary somatosensory cortex Neuroimage 10 63 83 10385582 32 Geyer S 1996 Two different areas within the primary motor cortex of man Nature 382 805 807 8752272 33 Zilles K Amunts K 2010 Centenary of Brodmann’s map–Conception and fate Nat Rev Neurosci 11 139 145 20046193 34 Amunts K Schleicher A Zilles K 2007 Cytoarchitecture of the cerebral cortex–More than localization Neuroimage 37 1061 1065 17870622 35 Zilles K Schleicher A Palomero-Gallagher N Amunts K 2002 Quantitative analysis of cyto-and receptor architecture of the human brain Brain Mapping: The Methods Elsevier San Diego 2nd Ed 573 602 36 Mahon BZ Caramazza A 2011 What drives the organization of object knowledge in the brain? Trends Cogn Sci 15 97 103 21317022 37 Heimler B Striem-Amit E Amedi A 2015 Origins of task-specific sensory-independent organization in the visual and auditory brain: Neuroscience evidence, open questions and clinical implications Curr Opin Neurobiol 35 169 177 26469211 38 Thaler L Goodale MA 2016 Echolocation in humans: An overview Wiley Interdiscip Rev Cogn Sci 7 382 393 27538733 39 Bi Y Wang X Caramazza A 2016 Object domain and modality in the ventral visual pathway Trends Cogn Sci 20 282 290 26944219 40 Ricciardi E Handjaras G Pietrini P 2014 The blind brain: How (lack of) vision shapes the morphological and functional architecture of the human brain Exp Biol Med (Maywood) 239 1414 1420 24962172 41 Bock AS Fine I 2014 Anatomical and functional plasticity in early blind individuals and the mixture of experts architecture Front Hum Neurosci 8 971 25566016 42 Renier L De Volder AG Rauschecker JP 2014 Cortical plasticity and preserved function in early blindness Neurosci Biobehav Rev 41 53 63 23453908 43 Ptito M Matteau I Gjedde A Kupers R 2009 Recruitment of the middle temporal area by tactile motion in congenital blindness Neuroreport 20 543 547 19240660 44 Lomber SG 2017 What is the function of auditory cortex when it develops in the absence of acoustic input? Cogn Dev 42 49 61 45 Bola Ł 2017 Task-specific reorganization of the auditory cortex in deaf humans Proc Natl Acad Sci USA 114 E600 E609 28069964 46 Striem-Amit E Vannuscorps G Caramazza A 2017 Sensorimotor-independent development of hands and tools selectivity in the visual cortex Proc Natl Acad Sci USA 114 4787 4792 28416679 47 Bedny M 2017 Evidence from blindness for a cognitively pluripotent cortex Trends Cogn Sci 21 637 648 28821345 48 Amedi A Hofstetter S Maidenbaum S Heimler B 2017 Task selectivity as a comprehensive principle for brain organization Trends Cogn Sci 21 307 310 28385460 49 Campus C Sandini G Concetta Morrone M Gori M 2017 Spatial localization of sound elicits early responses from occipital visual cortex in humans Sci Rep 7 10415 28874681 50 Striem-Amit E 2016 Topographical functional connectivity patterns exist in the congenitally, prelingually deaf Sci Rep 6 29375 27427158 51 Striem-Amit E 2015 Functional connectivity of visual cortex in the blind follows retinotopic organization principles Brain 138 1679 1695 25869851 52 Bock AS 2015 Resting-state retinotopic organization in the absence of retinal input and visual experience J Neurosci 35 12366 12382 26354906 53 Schieber MH 2001 Constraints on somatotopic organization in the primary motor cortex J Neurophysiol 86 2125 2143 11698506 54 Ramachandran VS Rogers-Ramachandran D Stewart M 1992 Perceptual correlates of massive cortical reorganization Science 258 1159 1160 1439826 55 Rauschecker JP 1999 Auditory cortical plasticity: A comparison with other sensory systems Trends Neurosci 22 74 80 10092047 56 Kaas JH 1990 Reorganization of retinotopic cortical maps in adult mammals after lesions of the retina Science 248 229 231 2326637 57 Karni A 1995 Functional MRI evidence for adult motor cortex plasticity during motor skill learning Nature 377 155 158 7675082 58 Nudo RJ Milliken GW Jenkins WM Merzenich MM 1996 Use-dependent alterations of movement representations in primary motor cortex of adult squirrel monkeys J Neurosci 16 785 807 8551360 59 Janssen P Scherberger H 2015 Visual guidance in control of grasping Annu Rev Neurosci 38 69 86 25840009 60 Martel M Cardinali L Roy AC Farnè A 2016 Tool-use: An open window into body representation and its plasticity Cogn Neuropsychol 33 82 101 27315277 61 Goldenberg G Spatt J 2009 The neural basis of tool use Brain 132 1645 1655 19351777 62 Tarhan LY Watson CE Buxbaum LJ 2015 Shared and distinct neuroanatomic regions critical for tool-related action production and recognition: Evidence from 131 left-hemisphere stroke patients J Cogn Neurosci 27 2491 2511 26351989 63 Caspers S Zilles K Laird AR Eickhoff SB 2010 ALE meta-analysis of action observation and imitation in the human brain Neuroimage 50 1148 1167 20056149 64 Tunik E Rice NJ Hamilton A Grafton ST 2007 Beyond grasping: Representation of action in human anterior intraparietal sulcus Neuroimage 36 T77 T86 17499173 65 Leshinskaya A Caramazza A 2015 Abstract categories of functions in anterior parietal lobe Neuropsychologia 76 27 40 25592369 66 Gallivan JP McLean DA Flanagan JR Culham JC 2013 Where one hand meets the other: Limb-specific and action-dependent movement plans decoded from preparatory signals in single human frontoparietal brain areas J Neurosci 33 1991 2008 23365237 67 Gallivan JP McLean DA Smith FW Culham JC 2011 Decoding effector-dependent and effector-independent movement intentions from human parieto-frontal brain activity J Neurosci 31 17149 17168 22114283 68 Haar S Dinstein I Shelef I Donchin O 2017 Effector-invariant movement encoding in the human motor system J Neurosci 37 9054 9063 28821649 69 Ogawa K Imai F 2016 Hand-independent representation of tool-use pantomimes in the left anterior intraparietal cortex Exp Brain Res 234 3677 3687 27591782 70 Heed T Leone FTM Toni I Medendorp WP 2016 Functional versus effector-specific organization of the human posterior parietal cortex: Revisited J Neurophysiol 116 1885 1899 27466132 71 Jacobs S Danielmeier C Frey SH 2010 Human anterior intraparietal and ventral premotor cortices support representations of grasping with the hand or a novel tool J Cogn Neurosci 22 2594 2608 19925200 72 Lewis JW 2006 Cortical networks related to human use of tools Neuroscientist 12 211 231 16684967 73 Gallivan JP Cavina-Pratesi C Culham JC 2009 Is that within reach? fMRI reveals that the human superior parieto-occipital cortex encodes objects reachable by the hand J Neurosci 29 4381 4391 19357266 74 Peeters R 2009 The representation of tool use in humans and monkeys: Common and uniquely human features J Neurosci 29 11523 11539 19759300 75 Talairach J Tournoux P 1988 Co-Planar Stereotaxic Atlas of the Human Brain Thieme New York 76 Friston KJ Worsley KJ Frackowiak RSJ Mazziotta JC Evans AC 1994 Assessing the significance of focal activations using their spatial extent Hum Brain Mapp 1 210 220 24578041 77 Friston KJ Holmes AP Worsley KJ 1999 How many subjects constitute a study? Neuroimage 10 1 5 10385576 78 Eickhoff SB 2005 A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data Neuroimage 25 1325 1335 15850749 79 Jeffreys H 1998 The Theory of Probability Oxford Univ Press Oxford 80 Rouder JN Speckman PL Sun D Morey RD Iverson G 2009 Bayesian t tests for accepting and rejecting the null hypothesis Psychon Bull Rev 16 225 237 19293088 81 Biswal B Yetkin FZ Haughton VM Hyde JS 1995 Functional connectivity in the motor cortex of resting human brain using echo-planar MRI Magn Reson Med 34 537 541 8524021 82 Glasser MF 2016 A multi-modal parcellation of human cerebral cortex Nature 536 171 178 27437579 83 Meier JD Aflalo TN Kastner S Graziano MS 2008 Complex organization of human primary motor cortex: A high-resolution fMRI study J Neurophysiol 100 1800 1812 18684903 84 Strother L Medendorp WP Coros AM Vilis T 2012 Double representation of the wrist and elbow in human motor cortex Eur J Neurosci 36 3291 3298 22845758