Humans possess the remarkable ability to project tactile sensations outside their body and onto a hand-held tool that they are using. Despite nearly a century of research, the computations underlying this projection have not been adequately addressed. In the present study, we used model-driven psychophysics to fill this gap. We hypothesized that tool-based sensory projection involves the remapping of touch from sensory feedback in the hand into an egocentric coordinate system. We first formalized the computational steps underlying tactile remapping. We designed a novel tool-sensing experiment that allowed us to rigorously test this model. In this task, participants contacted an object with a hand-held rod and then judged whether the object was above or below where they were currently looking. This comparison would only be possible if the touch on the tool was projected outside the hand. Crucially, both object location and gaze position varied independently, allowing us to characterize the hand-to-space-to-gaze remapping process. Model-based curve fitting provided strong evidence that all participants in our task projected touch outside their body and into gaze-centered coordinates. Crucially, the resolution of this projection was similar to what has been found for touch on the body. These findings provide the first step toward characterizing the computations underlying the spatial projection of touch on external objects, highlighting the incredible versatility of the sensorimotor system.NEW & NOTEWORTHY Humans possess the remarkable ability to project tactile sensations outside their body and onto a hand-held tool. Despite nearly a century of research, the computations underlying this projection have not been adequately addressed. We combined computational modeling and psychophysics to demonstrate that egocentric tactile remapping underlies tool-based sensory projection. These findings provide the first characterization of the computations underlying the spatial projection of touch on external objects, highlighting the incredible versatility of the sensorimotor system.

The ability to localize touch on the skin is a fundamental perceptual skill. Recent results show that this ability extends to hand-held tools, allowing participants to precisely localize along a rod's length. This tool-based localization may involve repurposing body-based somatosensory mechanisms. It remains unclear whether precise localization is limited to one-dimensional tools aligned with the long axis of the arm. Here, we show that touch can be localized along multiple dimensions of a held object. Participants held a square board while we measured tactile localization performance. Accuracy was well above chance, both overall and for both the proximodistal and medio-lateral axes. Notably, localization was more precise in the medio-lateral axis, reflecting an anisotropy similar to tactile perception on the skin. This anisotropy was defined in a hand-centered reference frame. These results show that wielded objects can be integrated with the somatosensory system, allowing rich tactile perception from tools.

The brain computes the spatiotopic position of touch by integrating tactile and proprioceptive signals (i.e., tactile remapping). While it is often assumed that the spatiotopic touch location is mapped into extrinsic, limb-independent coordinates, an alternative view proposes that touch is remapped into intrinsic, limb-specific coordinates. To test between these hypotheses, we used electroencephalography (EEG) and a novel tactile stimulation paradigm in which participants (N = 20, 19 females) received touch on their hands positioned at various locations relative to the body. Previous findings suggest that neural activity in primate sensorimotor and parietal regions monotonically encodes limb position, with their sustained firing rates increasing or decreasing across the workspace. These amplitude gradients, detectable at the population level in somatosensory evoked potentials, can be used to test predictions from each spatiotopic coding scheme. If touch is coded extrinsically, neural gradients should reflect changes of the external stimulus location, regardless of the limb. If coded intrinsically, gradients should be tied to the position of each limb and mirror each other between hands. Both univariate and multivariate EEG analyses found no evidence for extrinsic coding. Instead, we observed neural signatures of limb-specific, intrinsic spatiotopic coding, with the earliest emerging ∼160 ms after touch in centroparietal channels, later shifting to frontotemporal and parieto-occipital channels. Furthermore, a population-based neural network model of tactile remapping successfully reproduced the observed gradient patterns. These results show that the human brain localizes touch using an intrinsic, limb-specific spatial code, challenging the dominant assumption of extrinsic encoding in tactile remapping.

Disturbances in body perception are a consistent feature of Restrictive Eating Disorders (REDs). In adult patients diagnosed with REDs, body dissatisfaction and overestimation of body weight have been associated with altered tactile acuity and biases in proprioceptive localization. Whether similar alterations are present in adolescents-when the disorder often emerges-remains unclear. This study aimed to characterize possible alterations in tactile and body perception in adolescents diagnosed with REDs. Patients (N = 48) and age-matched healthy controls (HCs, N = 45) completed tasks measuring tactile acuity and body landmark localization accuracy on the hand and abdomen. Body dissatisfaction was also measured with clinical questionnaires. The REDs group showed greater bias in abdomen perception, whereas hand perception was similar to that of HCs. No differences in tactile processing emerged between REDs and HCs. The relations between perceptual biases, tactile acuity, and self-reported body dissatisfaction were not statistically significant. Our results suggest that biases in body perception in REDs are present in adolescence, but limited to specific body parts, such as the abdomen.

Body representations arise from the integration of multiple sensory signals that provide information about the body’s size, posture, and internal states. Vision often plays a dominant role, calibrating and aligning signals from touch, proprioception, and interoception. Yet, individuals born without sight still construct coherent and functional body representations, suggesting that the role of vision in building these representations may be more flexible than previously thought. Here, we synthesize recent behavioral and neuroimaging findings on body representations in blind individuals. We argue that blindness offers a powerful case study in how body representations can emerge through non-visual pathways, revealing the brain’s capacity to flexibly reorganize the sensory inputs supporting its internal models of the body. In doing so, we highlight the broader relevance of visual loss for understanding the plastic and multisensory nature of body representation.

In this Introduction, we have the pleasure of introducing the twelve articles of this Special Issue of Multisensory Research celebrating the life and works of Vincent Hayward. Vincent was a prolific scientist, collaborator, and colleague. As you will see by the variety of contributed papers, his influence spanned several fields and topics; from engineering to neurophysiology; from skin mechanics to olfactory metacognition. We and many others had the pleasure of knowing and working with Vincent. His boundless curiosity shines through in the papers of this Special Issue, and hopefully in our Introduction as well. Though gone, he is not forgotten; His legacy and influence lives on in the hearts and minds of colleagues studying the (neuro)science of body perception.

How does the brain integrate artificial body extensions into its somatosensory representation? While prior work has shown that tool use alters body representation, little is known about how artificial augmentations alter body representations as they are worn and used. Here, we investigated the dynamics of somatosensory plasticity using a custom-built exoskeletal device that extended users’ fingers by 10 cm. Across four time points, before, during (pre- and post-use), and after exoskeleton wear, participants completed a high-density proprioceptive mapping task. We observed three distinct phases of plasticity. First, simply wearing the exoskeleton led to a contraction of the perceived length of the biological finger. Second, following active use, both biological and artificial finger representations expanded significantly, an effect absent when participants trained with a non-augmenting control device. Third, a lasting aftereffect on biological finger representation was observed even after device removal. Our findings demonstrate that wearable augmentations are rapidly integrated into the body representation, with dynamic adjustments in proprioceptive space shaped by both structural and functional properties of the device. This work advances our understanding of how the sensorimotor system accommodates artificial extensions and highlights the potential for body- augmenting technologies to be intuitively integrated into body representation. These results have direct implications for the design of prosthetics, exoskeletons, and other assistive technologies aimed at extending human physical capacity. Significance statement How do wearable augmentations reshape our sense of the body? Using a custom- built exoskeleton that extended the fingers, we tracked changes in body representation across four stages of wear and use. We found that the sensorimotor system rapidly integrates artificial extensions into its somatosensory map: first contracting, then expanding perceived finger length in response to structure and function, with the shifts persisting after device removal. Our findings reveal dynamic, use-dependent plasticity in body representation, which has direct relevance for designing prosthetics, exoskeletons, and assistive devices. ### Competing Interest Statement The authors have declared no competing interest. European Research Council, https://ror.org/0472cxd90, 101076991

The brain localizes touch in space by integrating tactile and proprioceptive signals, a process known as tactile remapping. While it is often assumed that the remapped touch is encoded in an extrinsic, limb-independent reference frame, an alternative view proposes that touch may instead be represented within an intrinsic, limb-specific coordinate system. To test these hypotheses, we used electroencephalography (EEG) and a novel tactile stimulation paradigm in which participants received touch on their hands positioned at various locations relative to the body. Previous findings suggest that neural activity in primate sensorimotor and parietal regions monotonically encodes limb position. We therefore analyzed amplitude gradients in somatosensory evoked potentials (SEPs) to test predictions from each coding scheme. If touch is coded extrinsically, neural gradients should reflect changes of the external stimulus location, regardless of the limb. If coded intrinsically, gradients should be tied to the position of each limb and mirror each other between hands. Both univariate and multivariate EEG analyses found no evidence for extrinsic coding. Instead, we observed neural signatures of limb-specific, intrinsic spatial codes, the earliest emerging about 160 ms after touch in centro-parietal regions, later shifting to fronto-temporal and parieto-occipital areas. Furthermore, a population-based neural network model of tactile remapping successfully reproduced the observed gradient patterns. These results show that the human brain localizes touch using an intrinsic, limb-specific spatial code, challenging the dominant assumption of extrinsic encoding in tactile remapping. ### Competing Interest Statement The authors have declared no competing interest. NWO, NWO-SGW-406.21.GO.009, Interreg NWE-RE:HOME, NWA-ORC-1292.19.298 ERC, ERC 101076991 SOMATOGPS

When a mosquito lands on your finger, swatting it away requires your brain to calculate its location in the external space, which depends on the body’s 3D posture. Two competing hypotheses explain how the brain solves this challenge: the integration hypothesis , where tactile signals are transformed into spatial coordinates by integrating touch and posture information; and the cueing hypothesis , where touch merely cues a location on the body whose position is specified via proprioception. Adjudicating between these hypotheses is nearly impossible without modeling the latent factors underlying somatosensory spatial perception. We fill this gap in the present study. We first formalized each hypothesis from a Bayesian perspective: If touch merely triggers proprioceptive localization (cueing hypothesis), tactile and proprioceptive localization should rely on the same Bayesian computations, with identical prior expectations about the mosquito’s spatial location; If they involve distinct Bayesian computational processes (integration hypothesis), distinct prior expectations may shape tactile and proprioceptive localization. To test these predictions, we had nineteen participants localize either proprioceptive or tactile targets on their fingertips. We then fit their data with several Bayesian models of each hypothesis. Models allowing different prior distributions between modalities provided the best fit for most participants, with 17 out of 19 participants showing significantly different prior distributions across modalities. These provide strong computational evidence that tactile and proprioceptive localization rely on distinct computational mechanisms, a conclusion that has important implications for how we understand these everyday behaviors and their neural mechanisms. Author Summary When a mosquito lands on your finger and you swat it away, your brain must solve a challenging problem: determining where the mosquito is in space based on where it touched your skin and where your finger is positioned. Scientists have debated how the brain accomplishes this. One hypothesis proposes that the brain transforms touch signals by combining them with information about body posture—an integration process. An alternative hypothesis suggests that touch simply signals which body part was contacted, and the brain then locates that body part with only proprioception—essentially treating touch as only a cue. While these hypotheses make different predictions, distinguishing between them using behavior alone has proven difficult because the underlying computations remain hidden. We addressed this by having participants locate either touches on their fingertips or the fingertips themselves, then used Bayesian computational modeling to reveal the spatial expectations guiding these judgments. Our models showed that tactile and proprioceptive localization rely on distinct spatial expectations, with 17 of 19 participants showing significantly different patterns. These findings provide computational evidence that localizing touch involves transformations beyond simply locating the body, supporting the integration hypothesis and challenging the idea that touch merely cues body location. ### Competing Interest Statement The authors have declared no competing interest.

Current models of mental body representations (MBRs) indicate that tactile inputs feed some of them for different functions, implying that altering tactile inputs may affect mental body representations differently. Here, we tested this hypothesis by leveraging repetitive somatosensory stimulation (RSS), known to improve tactile perception by modulating primary somatosensory cortex (SI) activity, and measured its effects over the body image, body model and superficial schema in a randomized sham-controlled, double-blind crossover study. Results show that repetitive somatosensory stimulation affected the body image, participants perceiving their finger size as being smaller after repetitive somatosensory stimulation. While previous work showed an increase in finger size perception after tactile anaesthesia, these findings reveal that tactile inputs can diametrically modulate the body image. In contrast, repetitive somatosensory stimulation did not seem to alter the body model or superficial schema. In addition, we report a novel mislocalization pattern, with a bias towards the middle finger in the distal phalanges that reverses towards the thumb in the proximal phalanx, enriching the known distortions of the superficial schema. Overall, these findings provide novel insights into the functional organization of mental body representations and their relationships with somatosensory information. Reducing the perceived body size through repetitive somatosensory stimulation could be useful in helping treat body image disturbances.

One of the most consistent findings in the study of body representation is that they are geometrically distorted. This finding is in contrast with human’s dexterity and fine motor skills, which are thought to be computationally optimal. How do we achieve optimal motor control given distorted body representations? This question, framed as the hand paradox, is still unresolved. While several solutions have been proposed at a conceptual level, none have addressed the paradox in the context of the actual computations performed by the sensorimotor system during motor control. In the present article, we propose that the hand paradox is solved by the closed loop nature of motor control, where body state estimates are an optimal combination of sensory information and internal predictions. We first formalize a dynamic Bayesian model of body representation during optimal feedback control. We then perform several simulations using this model to show how representations of finger geometry, which are initially distorted, become accurate very quickly after movement onset. This undistortion occurs regardless of initial levels of distortion and is incomplete in only the most unlikely conditions. These model simulations suggest that the closed-loop nature of sensorimotor control may be the key to resolve the hand paradox.

Peripersonal space (PPS) - the immediate space surrounding our body - is known to enhance perceptual and neural processes. However, whether PPS influences voluntary action initiation and its temporal awareness remains underexplored. In this study, we tested whether spatial proximity per se affects the initiation of voluntary actions and the temporal awareness of associated events - from decision to action execution. Using an adapted Libet clock paradigm within a virtual reality environment, participants were presented with a clock either within their PPS (Near) or outside it (Far). They were required to initiate spontaneous actions and either report the perceived timing of their decision to act, report the timing of their action execution, or make no report. Results showed that when participants were not required to estimate timings, they initiated their actions earlier when the Libet clock was within PPS compared to when it was outside PPS. This suggests that there is an urgency induced by spatial proximity, which may be reduced by cognitive demands. Additionally, when the clock was within PPS, participants perceived their action decision and execution as occurring earlier than when it was outside PPS. By revealing that spatial proximity objectively modulates action urgency and subjectively influences temporal awareness of actions, these findings disclose a critical role of PPS in voluntary decision-making and temporal awareness. These results highlight a previously overlooked role of distance in decision-making processes and call for further experiments to understand how proximity can shape our experience of agency and control over actions.

It has been suggested that our brain re-uses body-based computations to localize touch on tools, but the neural implementation of this process remains unclear. Neural oscillations in the alpha and beta frequency bands are known to map touch on the body in external and skin-centered coordinates, respectively. Here, we pinpointed the role of these oscillations during tool-extended sensing by delivering tactile stimuli to either participants’ hands or the tips of hand-held rods. To disentangle brain responses related to each coordinate system, we had participants’ hands/tool tips crossed or uncrossed at their body midline. We found that midline crossing modulated alpha (but not beta) band activity similarly for hands and tools, also involving a similar network of cortical regions. Our findings strongly suggest that the brain uses similar oscillatory mechanisms for mapping touch on the body and tools, supporting the idea that body-based neural processes are repurposed for tool use.

To configure our limbs in space, the brain must compute their position based on sensory information provided by mechanoreceptors in the skin, muscles, and joints. Because this information is corrupted by noise, the brain is thought to process it probabilistically and integrate it with prior belief about arm posture, following Bayes' rule. Here, we combined computational modeling with behavioral experimentation to test this hypothesis. The model conceives the perception of arm posture as the combination of a probabilistic kinematic chain composed by the shoulder, elbow, and wrist angles, compromised with additive Gaussian noise, with a Gaussian prior about these joint angles. We tested whether the model explains errors in a virtual reality (VR)-based posture matching task better than a model that assumes a uniform prior. Human participants (N = 20) were required to align their unseen right arm to a target posture, presented as a visual configuration of the arm in the horizontal plane. Results show idiosyncratic biases in how participants matched their unseen arm to the target posture. We used maximum likelihood estimation to fit the Bayesian model to these observations and estimate key parameters including the prior means and its variance-covariance structure. The Bayesian model including a Gaussian prior explained the response biases and variance much better than a model with a uniform prior. The prior varied across participants, consistent with the idiosyncrasies in arm posture perception and in alignment with previous behavioral research. Our work clarifies the biases in arm posture perception within a new perspective on the nature of proprioceptive computations.NEW & NOTEWORTHY We modeled the perception of arm posture as a Bayesian computation. A VR posture-matching task was used to empirically test this Bayesian model. The Bayesian model including a nonuniform postural prior well explained individual participants' biases in arm posture matching.

To sense and interact with objects in the environment, we effortlessly configure our fingertips at desired locations. It is therefore reasonable to assume that the underlying control mechanisms rely on accurate knowledge about the structure and spatial dimensions of our hand and fingers. This intuition, however, is challenged by years of research showing drastic biases in the perception of finger geometry.1,2,3,4,5 This perceptual bias has been taken as evidence that the brain’s internal representation of the body’s geometry is distorted,6 leading to an apparent paradox regarding the skillfulness of our actions.7 Here, we propose an alternative explanation of the biases in hand perception—they are the result of the Bayesian integration of noisy, but unbiased, somatosensory signals about finger geometry and posture. To address this hypothesis, we combined Bayesian reverse engineering with behavioral experimentation on joint and fingertip localization of the index finger. We modeled the Bayesian integration either in sensory or in space-based coordinates, showing that the latter model variant led to biases in finger perception despite accurate representation of finger length. Behavioral measures of joint and fingertip localization responses showed similar biases, which were well fitted by the space-based, but not the sensory-based, model variant. The space-based model variant also outperformed a distorted hand model with built-in geometric biases. In total, our results suggest that perceptual distortions of finger geometry do not reflect a distorted hand model but originate from near-optimal Bayesian inference on somatosensory signals.

Tools can extend the sense of touch beyond the body, allowing the user to extract sensory information about distal objects in their environment. Though research on this topic has trickled in over the last few decades, little is known about the neurocomputational mechanisms of extended touch. In 2016, along with our late collaborator Vincent Hayward, we began a series of studies that attempted to fill this gap. We specifically focused on the ability to localize touch on the surface of a rod, as if it were part of the body. We have conducted eight behavioral experiments over the last several years, all of which have found that humans are incredibly accurate at tool-extended tactile localization. In the present article, we perform a model-driven re-analysis of these findings with an eye toward estimating the underlying parameters that map sensory input into spatial perception. This re-analysis revealed that users can almost perfectly localize touch on handheld tools. This raises the question of how humans can be so good at localizing touch on an inert noncorporeal object. The remainder of the paper focuses on three aspects of this process that occupied much of our collaboration with Vincent: the mechanical information used by participants for localization; the speed by which the nervous system can transform this information into a spatial percept; and whether body-based computations are repurposed for tool-extended touch. In all, these studies underscore the special relationship between bodies and tools.

The spatial limits of sensory acquisition (its sensory horizon) are a fundamental property of any sensorimotor system. In the present study, we sought to determine whether there is a sensory horizon for the human haptic modality. At first blush, it seems obvious that the haptic system is bounded by the space where the body can interact with the environment (e.g., the arm span). However, the human somatosensory system is exquisitely tuned to sensing with tools-blind-cane navigation being a classic example of this. The horizon of haptic perception therefore extends beyond body space, but to what extent is unknown. We first used neuromechanical modeling to determine the theoretical horizon, which we pinpointed as 6 m. We then used a psychophysical localization paradigm to behaviorally confirm that humans can haptically localize objects using a 6-m rod. This finding underscores the incredible flexibility of the brain's sensorimotor representations, as they can be adapted to sense an object many times longer than the user's own body.NEW & NOTEWORTHY There are often spatial limits to where an active sensory system can sample information from the environment. Hand-held tools can extend human haptic perception beyond the body, but the limits of this extension are unknown. We used theoretical modeling and psychophysics to determine these spatial limits. We find that the ability to spatially localize objects through a tool extends at least 6 m beyond the user's body.

It is often claimed that tools are embodied by their user, but whether the brain actually repurposes its body-based computations to perform similar tasks with tools is not known. A fundamental computation for localizing touch on the body is trilateration. Here, the location of touch on a limb is computed by integrating estimates of the distance between sensory input and its boundaries (e.g., elbow and wrist of the forearm). As evidence of this computational mechanism, tactile localization on a limb is most precise near its boundaries and lowest in the middle. Here, we show that the brain repurposes trilateration to localize touch on a tool, despite large differences in initial sensory input compared with touch on the body. In a large sample of participants, we found that localizing touch on a tool produced the signature of trilateration, with highest precision close to the base and tip of the tool. A computational model of trilateration provided a good fit to the observed localization behavior. To further demonstrate the computational plausibility of repurposing trilateration, we implemented it in a three-layer neural network that was based on principles of probabilistic population coding. This network determined hit location in tool-centered coordinates by using a tool’s unique pattern of vibrations when contacting an object. Simulations demonstrated the expected signature of trilateration, in line with the behavioral patterns. Our results have important implications for how trilateration may be implemented by somatosensory neural populations. We conclude that trilateration is likely a fundamental spatial computation that unifies limbs and tools.

Evolution has allowed humans to become proficient tool users, using tools to interact with their environment and functionally extend their limbs. Tools extend the sensorimotor boundaries of their user's body and in doing so modulate their body representations. In this chapter, we review the empirical evidence for this body-tool integration, focusing on the effects of basic tools as well as robotic limbs. We first explore how tools update the user's sensorimotor representations, reviewing behavioral and neural findings from the past decade, and novel research on development. Next, we focus on tool sensing, a new paradigm for investigating the ability to localize external tactile stimuli beyond the body. We then review how the sensorimotor system adapts to the use of robotic limbs, both in a medical and nonmedical context. Throughout the chapter, we also turn our attention to the past and the future, to discuss how evolution shaped our sensorimotor system and how humans now seem to be able to integrate robotic devices, further extending the limits of their sensorimotor system.