Embodied Motor Imagery reshapes body and space representations

Embodied Motor Imagery reshapes body and space representations | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Embodied Motor Imagery reshapes body and space representations Anne-Lise Jouen, Ahmad Kaddour, Florent Lebon, Peter Ford Dominey, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9343223/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Robotic telepresence offers a powerful framework for investigating embodiment and the plasticity of body and space representations. In this study, we examined how action observation (AO) and combined action observation with motor imagery (AOMI) modulate peripersonal space (PPS), body schema, and subjective embodiment when participants are immersed in a humanoid robot from a first-person perspective. Across two experiments using an identical telepresence setup, participants either passively observed robotic pointing movements or simultaneously observed and imagined performing the same actions. Behavioral measures assessed PPS boundaries and forearm metric representation, while questionnaires quantified ownership, self-location, and agency as markers of embodiment. Results showed that AO alone induced a contraction of PPS without altering body schema, accompanied by partial embodiment (i.e., characterized by ownership and self-location but weak agency). In contrast, AOMI produced additional changes in body schema and stronger embodiment, with enhanced agency and imagery vividness also correlating with self-location. These findings demonstrate a dissociation between spatial and bodily representations and highlight the critical role of intentional motor simulation in reshaping body schema during robotic embodiment. Overall, the study shows that embodiment in telepresence is a graded phenomenon in which motor imagery plays a key role in integrating robotic bodies into internal representations of the body self. Biological sciences/Neuroscience Biological sciences/Psychology Social science/Psychology Embodiment Robotic telepresence Motor imagery Action observation Body schema Peripersonal space Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Humans have the remarkable ability to embody external objects, tools, or even artificial agents, dynamically incorporating their visuo-spatial features into body schema and peripersonal space (PPS) representations. This phenomenon, commonly referred to as embodiment, relies on the integration of multisensory and motor-related processes that support the sense of owning a body, localizing oneself in space, and acting as an agent in the environment [1–3]. Such a visuo-spatial embodiment is supported by brain plasticity allowing the update of sensorimotor maps in response to environmental context and visuo-motor interactions. Both the body schema and PPS constitute core components of embodiment theory, as they form the functional interface between sensory experience and bodily action. Rather than being static representations, they dynamically structure the relationship between the self and the surrounding environment, notably by organizing visual space according to action possibilities [4]. Accordingly, embodiment entails a functional organization of visual space relative to the body and its action capabilities. Within this framework, visual space is typically distinguished into two main components: a near egocentric space (peripersonal), corresponding to hand-reaching actions, and a far retinocentric space (extrapersonal), extending beyond immediate reach and often associated with social interaction [5,6]. Supported by this dual organization, spatial encoding is flexible and context-dependent. For instance, studies in both animals and humans have shown that the boundaries of PPS can expand toward far space when using tools to reach distant objects [7–10]. In parallel, tool use can reshape the body schema itself by altering the perceived metrics of the body [11,12]. Such sensory–motor manipulations of body and space representations can also emerge during interactions with artificial or robotic effectors. For instance, synchronous visuo-motor movements with a detached robotic hand have been shown to shift the perceived location of one’s own hand toward the robot, even in the absence of tactile feedback or strong ownership sensations [13], suggesting that non-biological effectors can be partially incorporated into the body schema. In this context, robotic telepresence (i.e. setup in which a user remotely perceives and interacts with the environment through the sensors and effectors of a robotic surrogate) offers a unique framework to explore embodiment at a broader scale, defined as the conscious experience of the self as an acting body[14]. By providing a first-person visual perspective via head-mounted displays (HMD) and synchronized visual-motor feedback, robotic telepresence enables users to experience incorporation into a robot body. Previous work has shown that first-person visual and multisensory feedback can already induce whole-body embodiment of a walking humanoid robot surrogate, even with partial or delayed control of the robot’s movements[15]. This "beaming" into the robot can lead to a sense of ownership, agency, and self-location within the robot, even if its morphology differs from the human body[16]. Crucially, such embodied telepresence has been shown to increase robot acceptability and social closeness[17,18]. These effects suggest that embodiment is not only a perceptual illusion but also carries broader social and cognitive consequences. In the neuroscience perspective, embodiment is theorized to emerge from the integration of multisensory cues -including vision, proprioception, and touch- for the experience of ownership and location, along with intentional motor signals for the sense of agency. While passive visuo-motor synchrony can induce basic ownership illusions, as seen in the classic rubber hand illusion[19], the experience of agency requires internally generated motor commands, even if covert [20,21]. These principles are formalized in motor simulation theory[22,23], which posits that action-related neural processes can be internally activated in the absence of actual movement execution. Within this framework, Action Observation (AO) and Motor Imagery (MI) are understood as two complementary forms of covert motor simulation, capable of engaging motor networks without overt movement. AO operates as a bottom-up process, triggered by the observation of others' movements. It recruits the mirror neuron system, originally described in primates [24,25], and supports embodied functions such as imitation, intention understanding, and action prediction. AO is a widely recognized approach for enhancing motor skill instruction and accelerating the learning process [26,27]. Its effectiveness depends on factors such as the observer's attention to task-relevant features and the contextual relevance of the observed action [28,29]. In contrast, MI is a top-down process in which individuals consciously simulate an action, generating visual and kinesthetic representations without producing any overt movement[30]. During MI, individuals consciously simulate movements, triggering predictive models of their expected sensory consequences [31–33]. This simulation can support stronger embodiment by aligning predicted and observed outcomes. Importantly, MI and action execution recruit overlapping fronto-parietal circuits that are central to body schema and motor representations [22,34]. Because MI provides direct access to predicted sensory consequences, it is often regarded as the prototypical form of motor simulation and has been shown to enhance motor performance while promoting plastic changes within motor and body-related representations[35,36]. Recent studies highlight the particular effectiveness of combined AO and MI (AOMI), in which individuals simultaneously observe and mentally simulate the same action[37,38], in immersive and embodied settings, predominantly investigated in virtual reality contexts[39–41]. Virtual reality and robotic interfaces effectively provide first-person perspectives, realistic avatars, and synchronized feedback that enhance ownership, agency, and neural engagement [42–44]. Such designs have demonstrated improved motor performance, learning, and rehabilitation outcomes [45–47]. Notably, the efficacy of these immersive protocols appears to depend on the richness of sensory feedback, the degree of immersion, and the integration of intentional motor simulation. In the context of robotic embodiment, several studies have explored the use of MI within brain-computer interface (BCI) systems. For instance, Alimardani et al., [44] showed that imagining movements to control a humanlike android robot could elicit a strong sense of agency and embodiment in participants, even without physical execution. They further demonstrated that performance-related feedback enhanced both the subjective sense of embodiment and the effectiveness of MI in a short time, suggesting a reciprocal relationship between MI skill and embodiment. Similarly, Ono et al.,[45] developed a BCI-controlled robotic exoskeleton that provided proprioceptive feedback during hand-squeezing tasks, showing promising results for motor rehabilitation. These approaches underscore the potential of combining motor simulation and embodied robotic interaction to strengthen sensorimotor representations. While previous robotic studies have focused on active MI combined with BCI control and neurofeedback to induce embodiment[44,45], these paradigms involve deliberate interaction with the robotic effector. In contrast, our approach relies on passive telepresence based on AO or AOMI. We test whether internally generated motor simulation alone can modulate embodiment and body–space representations, even in the absence of explicit control or proprioceptive feedback. Indeed, despite these advances, the specific role of intentional motor signals in modulating body schema and PPS during robotic telepresence remains underexplored. While observation alone may recalibrate PPS by updating reachability estimates, reshaping the body schema to match a robotic effector may require intentional motor signals to remap limb and space metrics in an embodied model. Likewise, previous research suggests that, depending on the intentional content of the motor signal, agency over a virtual robotic limb can contract or expand PPS and body schema representations[48]. In the current study, we directly compare these mechanisms using two distinct experiments conducted in the same embodied telepresence setup: one based on Action Observation only (AO experiment), and a second based on combined Action Observation and Motor Imagery (AOMI experiment). Critically, this design enables the investigation of embodied MI, in which participants imagine actions while being physically embodied in a real, moving robotic body viewed from a first-person perspective. Unlike most AOMI protocols typically conducted offline or in virtual environments, MI here unfolds within the physical world, with real-time visual, spatial, and environmental feedback (i.e., seeing, hearing and interacting with the experimenter while embodied in the robot). To our knowledge, such a direct comparison between AO and AOMI within the same fully embodied telepresence context remains largely unexplored. We hypothesize that: i) both conditions will modulate PPS and body representations, reflecting recalibration driven by the robot’s visual perspective and movements; ii) AOMI will lead to differential effects on body schema metrics, with changes reflecting the contribution of intentional motor processes; iii) Embodiment experiences (including ownership, location, and agency) will be modulated by the task, with AOMI expected to promote stronger or more integrated embodiment compared to AO alone. By clarifying the contributions of these mechanisms, our study aims to advance the understanding of how motor simulation and 1st person perspective contribute to embodiment and plasticity of spatial/body representation during robotic telepresence. RESULTS Experiment 1 – Action Observation (AO) Behavioral measurements During the robotic pointing task in the AO condition (Fig. 1 ), the mean estimated position of the forearm bisection (BIS; behavioral measure of perceived forearm midpoint) was 19.50 cm (SD = 4.77) at pre-test and 18.48 cm (SD = 4.12) at post-test. This difference was not statistically significant (main effect of Time: F(1, 20) = 2.37, p = .14, η²p = .086), indicating no reliable change in perceived forearm midpoint following the robotic pointing task. In contrast, PPS (RD - reaching distance estimation, behavioral measure of perceived reachability boundary) changed significantly over time (main effect of Time: F(1, 20) = 13.09, p = .002, η²p = .38). As illustrated in Fig. 2 , the estimated PPS boundary decreased from pre-test (M = 0.61, SD = 0.09) to post-test (M = 0.57, SD = 0.10). Embodiment questionnaire For each of the three embodiment categories (OWN - Ownership, LOC – Location, AG - Agency; questionnaire-based measures of embodiment), the sensation of embodiment into the robot was significantly different from 0 (all p < .001). The ANOVA also revealed a significant main effect of Category (F(3, 66) = 6.13, p < .001, η²p = .22). Post-hoc tests showed that only OWN (p = .0022) and LOC (p = .0036) differed significantly from the Control (CTRL) category (Fig. 3 ). For each experiment, the Raincloud Plot displays Pre vs Post observations (dots) and boxplots (median line; hinges = 25th/75th percentiles; whiskers = extremes within 1.5 × IQR), as well as a representation of smoothed density. Values are expressed in meters (cm). For each experiment, the Raincloud Plot displays Pre vs Post observations (dots) and boxplots (median line; hinges = 25th/75th percentiles; whiskers = extremes within 1.5 × IQR), as well as a representation of smoothed density. Values are expressed in meters (m). Discussion Experience 1 - AO The results of Experiment 1 show that passive AO during embodied robotic telepresence is sufficient to induce selective modifications of spatial representations, as revealed by a significant contraction of PPS following the robotic pointing task. This effect likely reflects an updating of reachability estimates driven by the robot’s visual perspective and limb metrics. In contrast, no reliable change was observed in body schema as indexed by the forearm bisection task, suggesting that visual observation alone does not suffice to rescale internal representations of the body. At the subjective level, participants reported significant sensations of ownership and self-location within the robot, while agency remained comparatively weak. Together, these findings indicate that first-person visual exposure to a robotic body can support partial embodiment and spatial recalibration, but that the absence of intentional motor engagement limits both the emergence of agency and the plasticity of body schema representations. Taken together, these findings suggest that passive observation alone may be insufficient to modify internal body schema representations. We therefore asked whether introducing intentional motor simulation and enhancing agency through MI would produce additional changes. Experiment 2 was designed to address this question using an embodied AOMI condition. Experiment 2 – Action Observation + Motor Imagery (AOMI) Behavioral measurements During the robotic pointing task in the AOMI condition (Fig. 1 ), the estimated forearm bisection (BIS; behavioral measure of perceived forearm midpoint) shifted significantly toward the trunk from pre-test (M = 22.37 cm, SD = 3.63) to post-test (M = 19.83 cm, SD = 3.08). This change was statistically significant (main effect of Time: F(1, 20) = 11.27, p = .003, η²p = .36), indicating a contraction of the perceived forearm metric representation following the robotic pointing task. PPS (RD - reaching distance estimation, behavioral measure of perceived reachability boundary) also changed significantly over time (main effect of Time: F(1, 20) = 16.47, p < .001, η²p = .45). As illustrated in Fig. 2 , the estimated PPS boundary decreased from pre-test (M = 0.60, SD = 0.04) to post-test (M = 0.56, SD = 0.05). Embodiment questionnaire The AOMI experiment induced a significant sensation of embodiment, with scores significantly different from zero across categories (p < .001). The ANOVA revealed a significant main effect of Category (F(3, 60) = 19.42, p < .001, η²p = .49). Post-hoc tests showed that OWN (M = 37, SD = 18), LOC (M = 59, SD = 26), and AG (M = 41, SD = 25) were all significantly higher than the CTRL category (M = 0.14, SD = 15; p < .01). Additionally, LOC scores were higher than OWN (p = .002) and AG (p < .02). As shown in Fig. 3 , location and agency values were substantially higher in the AOMI condition than in AO alone. Correlations Between Motor Imagery and Embodiment questionnaires measures Correlation analyses examined the relationship between MI vividness ratings collected during the robotic pointing task and embodiment subscales (OWN, LOC, AG, and CTRL). Four Pearson correlations were computed. Two uncorrected correlations reached significance: AG–MI (r = .48, p = .031) and LOC–MI (r = .56, p = .010). Because multiple tests were performed, p-values were adjusted using the Bonferroni correction (α_adj = .0125). After correction, only the LOC–MI correlation remained significant (p = .010). Agency–MI (p = .031), Ownership–MI (r = .11, p = .630), and CTRL–MI (r = .17, p = .475) did not reach the corrected threshold. Discussion Experience 2 – AOMI Experiment 2 demonstrates that adding an explicit MI component during action observation in robotic telepresence qualitatively modifies the pattern of embodiment and its consequences on body representations. In contrast to AO alone, combined AOMI induced a significant shift in forearm midpoint estimates toward the trunk along with a reduction of PPS. This suggests that intentional motor simulation contributes to recalibrating both bodily and spatial metrics during embodied interaction with a robotic body. Subjective embodiment was also stronger and more differentiated in the AOMI condition, with ownership, self-location, and agency all exceeding control levels, and location emerging as the dominant dimension. Importantly, the correlation between MI vividness and self-location indicates that the degree of internal motor engagement directly relates to how strongly participants locate themselves within the robot. Together, these findings support the view that embodied MI provides a critical mechanism for reshaping internal body representations during robotic telepresence, beyond the effects of visual observation alone. Synthesis of Experiments 1 - AO and 2 - AOMI: Effects of Observation and Motor Imagery on Embodiment and Body Representation As summarized in Table 1 , the addition of MI to AO produced broader effects, encompassing body schema and agency, and underscoring a dissociation between PPS and body schema modulation. Table 1 Effects of action observation and motor imagery on peri-personal space, body schema, and embodiment Action Observation (Exp1) PPS (RD) Body Schema (BIS) Embodiment Reduced η²p = .38 Unchanged OWN, LOC Action Observation and Motor Imagery (Exp2) Reduced η²p = .45 Reduced η²p = .36 OWN, LOC, AG GENERAL DISCUSSION In this study, we used robotic telepresence to investigate how embodiment within a small-sized humanoid robot modulates spatial and bodily representations, specifically peripersonal space and body schema. Action observation (AO) alone induced a contraction of peripersonal space, whereas combining action observation with motor imagery (AOMI) additionally altered both peripersonal space and body schema. These findings indicate that embodiment-related plasticity depends not only on first-person visual perspective but also on intentional motor engagement. Thus, the nature of motor signals critically shapes how the body and surrounding space are internally represented during telepresence. These results can be interpreted in light of previous work demonstrating that tool use reshapes both peripersonal space and body schema, extending spatial representations and altering body metrics through sensorimotor interaction with the environment [7–12]. Consistent with this literature, the contraction of peripersonal space observed during AO likely reflects recalibrated reachability estimates based on the robot’s shorter arm metrics, similarly to tool-use paradigms in which new effectors are incorporated into spatial representations even without overt movement execution. In the present study, simply observing the movements of a shorter robotic limb from a first-person perspective was sufficient to induce changes in peripersonal space. Importantly, however, in this context, the moving robot arm cannot be reduced to the passive properties of a static instrument. Although the iCub robot’s arm appeared metallic, it functioned as an active effector rather than a tool, and immersive first-person telepresence likely promoted its interpretation as a surrogate body part This embodied integration may explain why passive observation alone modulated peripersonal space in our paradigm, whereas previous studies directly comparing active and passive tool use have shown that merely observing tool manipulation, without motor execution, is insufficient to induce reliable changes in body or spatial representations[49]. In contrast, adding MI during telepresence produced qualitatively stronger effects. Combined AO and MI induced not only peripersonal space contraction but also significant changes in body schema, reflected by a perceived shortening of the forearm segment. Thus, the AOMI condition did not simply amplify the observation effect; it introduced an additional functional component, suggesting that internally generated motor signals are necessary to update internal body metrics. In consequence, even in the absence of overt motor execution, the association of visual and intentional signals was sufficient to modify body representations, in addition to inducing space remapping. Mechanistically, this supports the idea that body schema remapping requires stronger sensorimotor involvement, particularly intentional motor signals[20,21,31,48]. During MI, internally generated motor commands help predict the robot’s movements and facilitate their integration into the body schema. Consistent with this interpretation, a similar dissociation has been reported in healthy aging, where reduced reliability of internal bodily signals leads to distortions of body representations while peripersonal space remains relatively preserved[50,51], further supporting the idea that body schema updating relies more strongly on internally generated sensorimotor signals than peripersonal space. At the subjective level, concurrent embodiment increased during AOMI compared with AO alone, particularly for the senses of self-location and agency. Similar embodiment phenomena have been reported in clinical contexts, such as post-stroke sensorimotor rehabilitation and in the integration and acceptance of prosthetic devices [52,53], where artificial limbs or surrogate body parts are progressively incorporated into the user’s body representation. This incorporation is thought to rely on sensorimotor mapping and visuo-motor congruence between real and artificial elements, allowing external effectors -including robotic limbs- to be integrated into internal body representations [16,17]. Together, our results indicate that visual immersion supports partial embodiment, whereas intentional motor engagement strengthens and deepens this incorporation. More broadly, our findings suggest that embodiment arises from both passive and active processes: passive visual observation can recalibrate peripersonal space and promote ownership and self-location, whereas the emergence of agency depends on intentional motor processing and the comparison between predicted and actual action outcomes[21,38]. The enhanced embodiment observed during AOMI can be interpreted in light of theoretical models of motor simulation. The outcome of the AOMI situation is theorized to enhance motor simulation by coupling bottom-up visual input with top-down MI signals, leading to greater activation of sensorimotor networks than either process alone[54,55]. Two complementary models explain AOMI’s efficacy: the Dual Action Simulation hypothesis[37,38,56], which suggests parallel motor representations for observed and imagined actions, and the Visual Guidance Hypothesis[57], which proposes that the imagined action is prioritized while AO serves primarily as a visual guide that primes and stabilizes MI. In our telepresence setup, participants simultaneously observed and imagined the robotic movements, a configuration consistent with AOMI principles that likely amplified sensorimotor engagement and the sense of embodiment. Importantly, our results extend the literature findings by demonstrating that AOMI in a physically embodied robotic context does not merely improve motor-related processes but can directly reshape both spatial and bodily representations. Furthermore, the stronger embodiment observed during AOMI—particularly the emergence of a sense of agency - and its correlation with imagery vividness suggests that motor simulation actively contributes to the construction of bodily self-representation. The greater sense of agency reported in the motor imagery condition is also consistent with previous AOMI research showing enhanced motor learning, stronger neural activation, and improved behavioral outcomes compared with action observation alone[54,55,58]. Moreover, immersive and embodied contexts such as virtual reality and robotics appear to further amplify these effects[41–43]. In terms of neural substrates, both MI and AO engage extensive motor-related brain networks, with substantial overlap between the two processes and motor execution. MI, in particular, involves widespread activation across the primary motor cortex (M1), supplementary motor area (SMA), premotor and parietal cortices, as well as prefrontal and subcortical structures[34,37,59–61]. A meta-analysis of fMRI studies identified a shared core network—including the premotor cortex, rostral parietal regions, and somatosensory areas—commonly activated during MI, AO, and overt action[62]. Interestingly, MI more closely resembles physical execution network due to its stronger subcortical involvement. In contrast, AO relies more heavily on visual input, activating shared motor representations via the mirror neuron system and engaging premotor and parietal regions involved in action understanding[59–61]. Importantly, studies have shown that combining AO and MI in mixed protocols enhances both neural activation and behavioral outcomes beyond what either technique achieves alone[37,54,55,58]. These previous findings provide a plausible neural substrate to account for the stronger embodiment effects observed in our study and highlight the added value of embodied AOMI protocols. Indeed, such approaches not only enhance motor performance but also reshape body schema and peripersonal space representations during telepresence, with important implications for teleoperation, rehabilitation, and assistive robotics. Beyond its neural underpinnings, the dissociation observed between body schema and peripersonal space in our study further highlights the role of intentional motor control in shaping embodiment. In the current study, only the combined sensory and motor signals triggered in the AOMI experiment induced transformations in both body schema and peripersonal space, accompanied by an additional sense of agency. To explain the link between motor processing and the modulation of body schema and peripersonal space, we suggest that higher-order intentional processes related to motor control are crucial for generating the sense of agency and the consequent changes in spatial and body representations. The sense of agency relies on the comparison between motor commands issued from intentional processes and the feedback signals informing the sensory consequences of an action. In the case of MI, although the motor command is attenuated to prevent overt execution, a copy of this motor signal is generated and compared with afferent signals arising from the observed visual scene[20,21,31]. Likewise, in our previous studies on human–robot interactions [16–18], we described how intentional motor commands triggered during active visuo-motor interactions were responsible for external agency and changes in social perception. During these robotic telepresence experiments, shared synchronous movements between the robot and the human may engage resonance mechanisms supported by the mirror neuron system, which are associated with several social functions, including theory of mind, empathy, and self-recognition[63,64]. Similar resonance mechanisms may also be elicited when one imagines generating a hand movement in synchrony with observing the robot’s movement. Based on these observations, we suggest that intentional commands underlying these simultaneous imagined or real movements are crucial for producing changes in body and spatial representations, as well as social feelings, co-occurring with the sensation of embodiment into the robot. This interpretation is consistent with theories of the mirror neuron network, which propose that resonance mechanisms in the brain allow shared action representations to emerge during robotic telepresence. CONCLUSION Our findings demonstrate that robotic telepresence can flexibly modulate both spatial and bodily representations in humans. More importantly, they reveal that robotic embodiment is a graded phenomenon shaped by the level of motor engagement. While visual observation alone is sufficient to recalibrate action-related space, the addition of motor imagery qualitatively extends this plasticity by engaging internal motor signals that contribute to body metric recalibration and strengthen the sense of agency. In this sense, AOMI does not merely amplify the effects of observation, but introduces an additional layer of embodiment, linking spatial remapping, bodily self-representation, and intentional motor processing within a unified framework. Beyond these experimental insights, embodied AOMI protocols in robotic telepresence hold strong basic and translational potential. By engaging motor networks without overt movement, they offer a powerful paradigm to investigate the neural foundations of bodily self-representation and space awareness in immersive real-world contexts. At the same time, such approaches provide a concrete framework for neurorehabilitation and assistive technologies, positioning embodiment within robotic or virtual avatars as a valuable bridge between fundamental neuroscience and therapeutic innovation. MATERIAL AND METHODS Experiment 1 – Action Observation (AO) - Participants A group of 21 healthy, right-handed participants (10 males, mean age = 24.38 years, SD = 6.5) participated in Experiment 1 – AO. All participants had normal or corrected-to-normal vision and reported no history of drug use, neurological, or psychiatric disorders. Participants were naïve to the purpose of the study and provided written informed consent prior to participation. The experimental protocol was approved by the Ethics Committee for Research of Université Bourgogne Europe (CER UBE; approval number: CERUBFC-2021-07-10-028) and conducted in accordance with the Declaration of Helsinki (1964). Individuals appearing in the figures provided informed consent for publication of identifying images in an online open-access format. - Apparatus and Setup The AO experiment employed a robotic telepresence setup in which participants wore a head-mounted display (HMD) providing a real-time, first-person visual perspective from the robot’s viewpoint. The study was conducted using the humanoid robot iCub (Metta et al., 2010), which is approximately the size of a 3-year-old child and equipped with video cameras in its eyes, enabling it to explore and perceive its visual environment. The stereo video signal from the two cameras is transmitted to the HMD (SONY HMZ-3WT 3D Viewer) to generate for the participant a stereo immersion in the scene from the robot’s perspective, including vision of the robot’s moving arms and hands, to induce a sense of embodiment into the robot body. Participants sat comfortably at a table in a quiet room, wearing the HMD connected to the robot’s camera system. The robot was placed elsewhere in the experimental room, out of the participant's view from the robot's perspective as illustrated in (Fig. 4 -a). During the telepresence tasks, the robot executed a series of preprogrammed pointing movements with its hand and forearm. The dimensions of the robotic limb were smaller than those of a typical human limb, but this mismatch was not explicitly mentioned by the experimenter. (a) Robotic telepresence setup: the participant wears a HMD and receives real-time visual input from the robot’s cameras. (b) Reaching distance estimation task used to assess PPS, with a red ball moving along a rail toward or away from the robot. (c) Robotic pointing task in which the robot points toward coloured cubes arranged on a grid. (d) Same robotic pointing task viewed from the robot first-person perspective as seen by the participant through the HMD during telepresence - Procedure After a general explanation of the tasks, the participant was familiarized with the iCub robot through a brief demonstration of the robot’s head and arm motor capabilities. The experimental procedure comprised five successive phases and lasted approximately 40 minutes in total (see Fig. 5 ). The forearm bisection (BIS) and reaching distance (RD) were used as behavioral measurements of body schema and PPS representations, respectively, whereas the robotic pointing task constituted the main experimental task. All phases but the BIS measurement took place with the HMD i.e., during telepresence. Participants first completed a BIS measurement without the HMD. Then, they wore the HMD and performed a RD measurement, followed by the robotic pointing task, during which they passively observed the robot’s movements from a first-person perspective (AO experiment). Still wearing the HMD, participants subsequently performed a second RD measurement to assess changes in PPS. Finally, the HMD was removed and a final BIS measurement was conducted to evaluate changes in body representation. The two experiments shared an identical structure and behavioural measurements, with the only difference being the inclusion of motor imagery evaluation/training and in-task ratings in Experiment 2. - Behavioral measurements o Forearm bisection (BIS) The BIS measure was used to assess changes in body schema, specifically the perceived metric representation of the arm. Participants’ right arm was positioned parallel to the midsagittal plane on the table and covered with a rigid screen to prevent visual and tactile feedback. While blindfolded, participants pointed with their left index finger to the perceived midpoint of their right forearm, defined as the segment between the elbow and the tip of the middle finger[48,65]. To minimize spatial prediction, the left hand was placed in different initial positions across trials. After three practice trials, participants completed 12 experimental trials with randomized starting positions. For each trial, the indicated midpoint was recorded using a graduated scale fixed to the rigid cover. o Reaching distance estimation (RD) The RD measure, adapted from D’Angelo et al.[48], was used to assess the boundaries of PPS. From the robot’s first-person perspective, viewed through the HMD, participants observed a table equipped with a 150-cm drylin® rail along which a red ball (8-cm diameter), mounted on a small platform, could move either toward (approaching trials) or away from (withdrawing trials) the robot (Fig. 4 -b). The ball’s starting position (25 and 100 cm for withdrawing and approaching trials, respectively) and speed (approximately 2.75 cm/s) were precisely controlled by the experimenter using a joystick. During each trial, participants verbally indicated, without moving, when the ball reached the boundary at which it was judged reachable (approaching trials) or unreachable (withdrawing trials), following the instruction: “Say stop when you estimate that you could or could not reach the ball.” The final stopping position was recorded using a laser telemeter measuring the distance between the robot’s midsagittal axis and the ball location at the stop position. Between two trials, participants briefly closed their eyes while the experimenter repositioned the ball for the next trial. After two practice trials, participants completed a total of 12 experimental trials, including six trials for each motion direction (toward and away from the robot). Both the BIS and RD measurements were administered twice, once before and once after the AO pointing task, using the same protocol. - Robotic pointing task In the robotic pointing task, participants wearing an HMD viewed the robot’s forearm from a first-person perspective as it pointed toward colored cubes arranged on a chessboard (Fig. 4 -c). A 75 × 60 cm tilted grid tablet was positioned in front of the robot and centered on its body axis. Prior to the experiment, the robot’s PPS boundary was objectively determined through preliminary reaching measurements to estimate the maximum reachable workspace of the robotic arm. This calibration was used to spatially classify targets as reachable (PPS) or unreachable (extrapersonal space EPS), resulting in an intentionally asymmetric distribution of cubes (5 within PPS and 7 beyond reach, in EPS) due to the smaller size of the robot’s reachable workspace. On the tablet, the 12 cubes (5 cm³) were arranged to cover the grid accordingly. At each trial, the robot pointed with its right hand to a cube that had been previously selected and lifted for 2 seconds by the experimenter as illustrated in Fig. 4 -c, d. While the robot hand was pointing to a cube, the experimenter verbally indicated whether the robot hand was touching the corresponding cube or not, and then returned the cube to its initial position. To optimize coordinated reaching movements and visual fixation, the robot’s motion was programmed using a combination of trunk rotation, forearm displacement, and head orientation (iCub ctpservice function) to ensure maximal accuracy across spatial positions. The sequences of pointing movements were pre-established, and the experimenter was cued before each trial regarding the target location. A video demonstrating this behavior can be seen: https://youtu.be/aYQlfk9EOVw . The pointing task consisted of 10 sequences of 10 pointing movements executed by the robot toward cubes located either within or beyond reaching distance. The order of cubes and sequences was randomized across participants to ensure a balanced number of reachable and non-reachable trials. Through the HMD, participants could clearly perceive the movements of the robot’s hand, fingers, and forearm (Fig. 4 -d). During the robotic pointing task, participants were instructed to carefully observe the spatial positions of the cubes, the robot’s hand and forearm movements, and whether each target was within or beyond the robot’s reachable space (Fig. 4 -d). They were asked to remain relaxed, refrain from any physical movement, and simply observe the robot’s pointing gestures from a first-person perspective through the HMD. No explicit instructions regarding embodiment or perspective-taking were provided, allowing any sensations of embodiment to emerge naturally. The session consisted of 10 sequences of 10 pointing movements (100 movements in total) and lasted approximately 10 minutes. - Embodiment questionnaire At the end of the experiment, participants were invited to complete a questionnaire aimed at quantifying their sensation of embodiment into the robot during the robotic pointing task. This questionnaire, consisting of 10 statements (see Supplementary Materials: SM1), allowed evaluation of embodiment on a subjective scale ranging from 0 (no sensation) to 100 (extremely strong sensation). For each statement, participants indicated a value between 0 and 100 specifying the level of embodiment felt during the experiment. The embodiment score was assessed across three categories of statements focusing on: (i) the sensation that the robot’s hand belonged to the participant (3 items), corresponding to ownership (OWN); (ii) the sensation that the participant’s hand was located at the position of the robot’s hand, or that they were located at the robot’s position (2 items), corresponding to self-location (LOC); and (iii) the sensation of controlling the robot’s movements (3 items), corresponding to agency (AG), reflecting the perceived motor control exerted over the robot. Two control statements, assumed to be independent of the experimental manipulation, were also included. Scores for each category were computed as the average of the ratings across the corresponding items. Data analysis Statistical analyses and figures were performed using Statistica (TIBCO Software Inc., CA 94304, USA) and JASP[66]. Data were first assessed for normality (Kolmogorov–Smirnov) and homogeneity of variances (Levene’s test). As all assumptions were met (p > .05), parametric tests were applied. The significance level was set at α = .05. Mixed repeated-measures ANOVAs were conducted for BIS (i.e., estimated midpoint of the hand–forearm segment) and RD (estimation of the PPS boundary) measures, with Time (Pre, Post) as a within-subject factor. Embodiment questionnaire scores were analyzed using repeated-measures ANOVAs with Category (Ownership OWN, Location LOC, Agency AG, Control CTRL) as a within-subject factor. Significant effects were followed by Bonferroni-corrected pairwise comparison. Experiment 2 – Action Observation + Motor Imagery (AOMI) - Participants A different group of 21 participants (8 males, mean age = 22.89 years, SD = 5.0) participated in Experiment 2 – AOMI. Inclusion criteria were identical to those described for Experiment 1 – AO. All participants provided written informed consent prior to participation. The study was approved by the Ethics Committee for Research of Université Bourgogne Europe (CER UBE; approval number: CERUBFC-2021-07-10-028) and conducted in accordance with the Declaration of Helsinki. - Apparatus and Setup The same robotic telepresence apparatus, HMD, and iCub humanoid robot described in Experiment 1 were used. - Procedure The overall procedure was identical to that of Experiment 1, including the sequence and timing of the behavioral measurements (BIS and RD) and the robotic pointing task (see Fig. 5 ). The only differences concerned the instructions given to the participants and the addition of a MI training and evaluation phase prior to telepresence exposure, as well as intermittent MI vividness ratings collected between blocks during the robotic pointing task. The total duration of the Experiment 2 - AOMI was approximately 45–50 minutes. - MI evaluation and training Prior to the behavioral measurements, participants’ intrinsic MI vividness was assessed using the KVIQ-10, a short version of the Kinesthetic and Visual Imagery Questionnaire (KVIQ-20; [67]). The KVIQ evaluates, on a five-point ordinal scale, the clarity of visual imagery and the intensity of kinesthetic sensations experienced during first-person MI. Participants were seated comfortably and performed 10 simple movements, each followed by the corresponding imagined movement (performed with eyes open), which they rated using both visual and kinesthetic subscales. For MI training, participants sat in front of a table on which three colored cubes were positioned approximately 40 cm away, reproducing the spatial configuration and viewing angle of the robotic pointing task. First, participants physically performed pointing movements toward each cube (four repetitions per cube, 12 trials total). Second, they were asked to imagine performing the same pointing gestures while focusing on both visual and kinesthetic sensations. After each imagined trial, participants rated the vividness of their imagery on a scale from 0 (no sensation) to 5 (as if actually moving), reflecting both kinesthetic and visual aspects. After this MI training phase, participants proceeded with the first BIS and RD measurements, as described in Experiment 1 - AO. - Behavioral measurements and embodiment questionnaire The BIS, RD, and embodiment questionnaire measures were administered and processed in the same manner as in Experiment 1 – AO. - Robotic pointing task Participants then performed the robotic pointing task. Using the same setup, they observed identical robotic pointing movements through the HMD for approximately 10 minutes, as in the AO experiment. The key difference in this condition was that they were explicitly instructed to engage in MI, imagining themselves performing the same pointing movements in real time as precisely as possible. During the task, they observed the robot’s hand and forearm, the spatial location of each cube, and whether it was reachable or not by the robot, while simultaneously imagining executing the same movement. They were asked to focus on the kinesthetic sensations, mentally simulating the feeling of their own arm and hand moving in synchrony with the robot. Each sequence included 10 robotic pointing movements and was repeated until all 10 movements were completed, for a total of 100 movements. At the end of each 10-movement sequence, participants rated their imagery vividness on a scale from 0 (no sensation at all) to 5 (as if actually moving), reflecting both kinesthetic and visual impressions, as in the earlier MI training. Data analysis Statistical analyses followed the same procedure as described for Experiment 1 – AO. Additional analyses were conducted to examine the MI component. KVIQ scores and imagery vividness ratings obtained during training indicated that participants had an average KVIQ total score of 30.90/50 and an average training vividness score of 3.22/5; these measures were therefore not further analyzed. MI vividness ratings collected during the robotic pointing task were used to verify that participants maintained a consistent level of imagery vividness throughout the session, thereby limiting potential mental fatigue[68]. These MI ratings were also correlated with embodiment questionnaire scores using Pearson’s correlation coefficient (r).</p Declarations Author contributions J.V.D. and. P.F.D conceived the original research idea. All authors contributed to the study conception and experimental design. A.K. and P.F.D. developed and implemented the robotic platform. A.L.J. conducted participant recruitment, and A.L.J. and J.V.D. carried out data collection and analyses. A.L.J. and J.V.D. drafted the initial manuscript. All authors contributed to the interpretation of the results and provided critical revisions of the manuscript. A.L.J. prepared the final version of the manuscript. All authors reviewed and approved the final manuscript and agreed to its submission to Scientific Reports . Data availability statement The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request. Additional Information (including a Competing Interests Statement) The authors declare no competing interests. Funding Declaration This research was supported by the French Region Bourgogne-Franche-Comté (Grant ANER RobotSelf 2019-Y-10650). References Gallagher, S. Dimensions of Embodiment: Body Image and Body Schema in Medical Contexts. in Handbook of Phenomenology and Medicine (ed. Toombs, S. K.) 147–175 (Springer Netherlands, Dordrecht, 2001). doi:10.1007/978-94-010-0536-4_8. de Vignemont, F. Embodiment, ownership and disownership. Consciousness and Cognition 20 , 82–93 (2011). Blanke, O. Multisensory brain mechanisms of bodily self-consciousness. Nat Rev Neurosci 13 , 556–571 (2012). Borghi, A. M. & Cimatti, F. Embodied cognition and beyond: Acting and sensing the body. Neuropsychologia 48 , 763–773 (2010). Berti, A. & Rizzolatti, G. 3.3 Coding near and far space. The cognitive and neural bases of spatial neglect 119 (2002). Berti, A. From embodying tool to embodying alien limb: sensory-motor modulation of personal and extrapersonal space. Cogn Process 22 , 121–126 (2021). Iriki, A., Tanaka, M. & Iwamura, Y. Coding of modified body schema during tool use by macaque postcentral neurones. Neuroreport 7 , 2325–2330 (1996). Berti, A. & Frassinetti, F. When Far Becomes Near: Remapping of Space by Tool Use. Journal of Cognitive Neuroscience 12 , 415–420 (2000). Maravita, A. & Iriki, A. Tools for the body (schema). Trends in Cognitive Sciences 8 , 79–86 (2004). Canzoneri, E. et al. Tool-use reshapes the boundaries of body and peripersonal space representations. Exp Brain Res 228 , 25–42 (2013). Cardinali, L. et al. Tool-use induces morphological updating of the body schema. Current biology 19 , R478–R479 (2009). Cardinali, L. et al. Grab an object with a tool and change your body: tool-use-dependent changes of body representation for action. Exp Brain Res 218 , 259–271 (2012). Romano, D., Caffa, E., Hernandez-Arieta, A., Brugger, P. & Maravita, A. The robot hand illusion: Inducing proprioceptive drift through visuo-motor congruency. Neuropsychologia 70 , 414–420 (2015). Garbarini, F. et al. When your arm becomes mine: pathological embodiment of alien limbs using tools modulates own body representation. Neuropsychologia 70 , 402–413 (2015). Aymerich-Franch, L., Petit, D., Ganesh, G. & Kheddar, A. Embodiment of a humanoid robot is preserved during partial and delayed control. in 2015 IEEE International Workshop on Advanced Robotics and its Social Impacts (ARSO) 1–5 (IEEE, 2015). Farizon, D., Dominey, P. F. & Ventre-Dominey, J. Insights on embodiment induced by visuo-tactile stimulation during robotic telepresence. Sci Rep 11 , 22718 (2021). Ventre-Dominey, J. et al. Embodiment into a robot increases its acceptability. Sci Rep 9 , 10083 (2019). Farizon, D., Dominey, P. F. & Ventre-Dominey, J. Social Attitude Towards a Robot is Promoted by Motor-Induced Embodiment Independently of Spatial Perspective. IEEE Robotics and Automation Letters 7 , 9036–9042 (2022). Botvinick, M. & Cohen, J. Rubber hands ‘feel’ touch that eyes see. Nature 391 , 756–756 (1998). Jeannerod, M. The representing brain: Neural correlates of motor intention and imagery. Behavioral and Brain sciences 17 , 187–202 (1994). Jeannerod, M. Mental imagery in the motor context. Neuropsychologia 33 , 1419–1432 (1995). Jeannerod, M. Neural simulation of action: a unifying mechanism for motor cognition. Neuroimage 14 , S103–S109 (2001). Jeannerod, M. Motor Cognition: What Actions Tell the Self . vol. 42 (OuP Oxford, 2006). di Pellegrino, G., Fadiga, L., Fogassi, L., Gallese, V. & Rizzolatti, G. Understanding motor events: a neurophysiological study. Exp Brain Res 91 , 176–180 (1992). Rizzolatti, G. & Craighero, L. THE MIRROR-NEURON SYSTEM. Annual Review of Neuroscience 27 , 169–192 (2004). Ste-Marie, D. M. et al. Observation interventions for motor skill learning and performance: an applied model for the use of observation. International Review of Sport and Exercise Psychology 5 , 145–176 (2012). Ste-Marie, D. M., Lelievre, N. & St. Germain, L. Revisiting the Applied Model for the Use of Observation: A Review of Articles Spanning 2011–2018. Research Quarterly for Exercise and Sport 91 , 594–617 (2020). Schuch, S., Bayliss, A. P., Klein, C. & Tipper, S. P. Attention modulates motor system activation during action observation: evidence for inhibitory rebound. Exp Brain Res 205 , 235–249 (2010). Wright, D. J. et al. Directing visual attention during action observation modulates corticospinal excitability. PLOS ONE 13 , e0190165 (2018). Decety, J. The neurophysiological basis of motor imagery. Behavioural brain research 77 , 45–52 (1996). Frak, V., Paulignan, Y. & Jeannerod, M. Orientation of the opposition axis in mentally simulated grasping. Experimental Brain Research 136 , 120–127 (2001). Jacobs, S., Danielmeier, C. & Frey, S. H. Human anterior intraparietal and ventral premotor cortices support representations of grasping with the hand or a novel tool. Journal of Cognitive Neuroscience 22 , 2594–2608 (2010). Rieger, M. & Massen, C. Tool characteristics in imagery of tool actions. Psychological Research 78 , 10–17 (2014). Lotze, M. & Halsband, U. Motor imagery. Journal of Physiology-paris 99 , 386–395 (2006). Ruffino, C., Papaxanthis, C. & Lebon, F. Neural plasticity during motor learning with motor imagery practice: Review and perspectives. Neuroscience 341 , 61–78 (2017). Moran, A. & O’Shea, H. Motor imagery practice and cognitive processes. Frontiers in Psychology 11 , 394 (2020). Eaves, D. L., Riach, M., Holmes, P. S. & Wright, D. J. Motor Imagery during Action Observation: A Brief Review of Evidence, Theory and Future Research Opportunities. Front. Neurosci. 10 , (2016). Vogt, S., Di Rienzo, F., Collet, C., Collins, A. & Guillot, A. Multiple roles of motor imagery during action observation. Front. Hum. Neurosci. 7 , (2013). Adamo, P. et al. Effects of Action Observation Plus Motor Imagery Administered by Immersive Virtual Reality on Hand Dexterity in Healthy Subjects. Bioengineering 11 , (2024). Connelly, N., Welsby, E., Lange, B. & Hordacre, B. Virtual Reality Action Observation and Motor Imagery to Enhance Neuroplastic Capacity in the Human Motor Cortex: A Pilot Double-blind, Randomized Cross-over Trial. Neuroscience 549 , 92–100 (2024). Frank, C. et al. Motor imagery during action observation in virtual reality: the impact of watching myself performing at a level I have not yet achieved. International Journal of Sport and Exercise Psychology 21 , 401–427 (2023). Choi, J. W., Kim, B. H., Huh, S. & Jo, S. Observing Actions Through Immersive Virtual Reality Enhances Motor Imagery Training. IEEE Transactions on Neural Systems and Rehabilitation Engineering 28 , 1614–1622 (2020). Lakshminarayanan, K. et al. The effect of combining action observation in virtual reality with kinesthetic motor imagery on cortical activity. Front. Neurosci. 17 , (2023). Alimardani, M., Nishio, S. & Ishiguro, H. The Importance of Visual Feedback Design in BCIs; from Embodiment to Motor Imagery Learning. PLOS ONE 11 , e0161945 (2016). Ono, Y., Wada, K., Kurata, M. & Seki, N. Enhancement of motor-imagery ability via combined action observation and motor-imagery training with proprioceptive neurofeedback. Neuropsychologia 114 , 134–142 (2018). Holper, L. et al. Testing the potential of a virtual reality neurorehabilitation system during performance of observation, imagery and imitation of motor actions recorded by wireless functional near-infrared spectroscopy (fNIRS). J NeuroEngineering Rehabil 7 , 57 (2010). Binks, J. et al. Combining Action Observation and Motor Imagery (AO+MI): Mechanisms, Applications, and Guidelines for Enhancing Motor Learning and Rehabilitation. in The Psychology of Mental Imagery in Movement and Action (Routledge, 2025). D’Angelo, M., Di Pellegrino, G., Seriani, S., Gallina, P. & Frassinetti, F. The sense of agency shapes body schema and peripersonal space. Scientific reports 8 , 13847 (2018). Galigani, M. et al. Effect of tool-use observation on metric body representation and peripersonal space. Neuropsychologia 148 , 107622 (2020). Costello, M. C. & Bloesch, E. K. Are older adults less embodied? A review of age effects through the lens of embodied cognition. Frontiers in psychology 8 , 267 (2017). Sorrentino, G. et al. How ageing shapes body and space representations: A comparison study between healthy young and older adults. Cortex 136 , 56–76 (2021). Verschure, P. F. M. J., Páscoa dos Santos, F. & Sharma, V. Redefining stroke rehabilitation: Mobilizing the embodied goal-oriented brain. Current Opinion in Neurobiology 83 , 102807 (2023). Zbinden, J., Lendaro, E. & Ortiz-Catalan, M. Prosthetic embodiment: systematic review on definitions, measures, and experimental paradigms. J NeuroEngineering Rehabil 19 , 37 (2022). Chye, S. et al. The effects of combined action observation and motor imagery on corticospinal excitability and movement outcomes: Two meta-analyses. Neuroscience & Biobehavioral Reviews 143 , 104911 (2022). The effect of action observation and motor imagery combinations on upper limb kinematics and EMG during dart‐throwing - Romano Smith - 2019 - Scandinavian Journal of Medicine & Science in Sports - Wiley Online Library. https://onlinelibrary.wiley.com/doi/abs/10.1111/sms.13534. Eaves, D. L., Haythornthwaite, L. & Vogt, S. Motor imagery during action observation modulates automatic imitation effects in rhythmical actions. Frontiers in human neuroscience 8 , 28 (2014). Meers, R., Nuttall, H. E. & Vogt, S. Motor imagery alone drives corticospinal excitability during concurrent action observation and motor imagery. Cortex 126 , 322–333 (2020). Scott, M. W. An error in our thinking: The sensory conflict hypothesis of combined action observation and motor imagery. Brain and Cognition 189 , 106350 (2025). Grèzes, J. & Decety, J. Functional anatomy of execution, mental simulation, observation, and verb generation of actions: A meta-analysis. Human Brain Mapping 12 , 1–19 (2001). Hétu, S. et al. The neural network of motor imagery: an ALE meta-analysis. Neuroscience & Biobehavioral Reviews 37 , 930–949 (2013). Caspers, S., Zilles, K., Laird, A. R. & Eickhoff, S. B. ALE meta-analysis of action observation and imitation in the human brain. NeuroImage 50 , 1148–1167 (2010). Hardwick, R. M., Caspers, S., Eickhoff, S. B. & Swinnen, S. P. Neural correlates of action: Comparing meta-analyses of imagery, observation, and execution. Neuroscience & Biobehavioral Reviews 94 , 31–44 (2018). Differentiating between self and others: an ALE meta-analysis of fMRI studies of self-recognition and theory of mind | Brain Imaging and Behavior | Springer Nature Link. https://link.springer.com/article/10.1007/s11682-013-9266-8. Gallese, V. Being like me”: Self-other identity, mirror neurons and empathy. Perspectives on imitation: From cognitive neuroscience to social science 1 , 101–18 (2005). Sposito, A. V., Bolognini, N., Vallar, G., Posteraro, L. & Maravita, A. The spatial encoding of body parts in patients with neglect and neurologically unimpaired participants. Neuropsychologia 48 , 334–340 (2010). Love, J. et al. JASP: Graphical statistical software for common statistical designs. Journal of Statistical Software 88 , 1–17 (2019). Malouin, F. et al. The Kinesthetic and Visual Imagery Questionnaire (KVIQ) for Assessing Motor Imagery in Persons with Physical Disabilities: A Reliability and Construct Validity Study. Journal of Neurologic Physical Therapy 31 , 20 (2007). Rozand, V., Lebon, F., Papaxanthis, C. & Lepers, R. Effect of mental fatigue on speed–accuracy trade-off. Neuroscience 297 , 219–230 (2015). Additional Declarations No competing interests reported. Supplementary Files SUPPLEMENTARYMATERIALS.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 29 Apr, 2026 Reviewers agreed at journal 28 Apr, 2026 Reviewers invited by journal 28 Apr, 2026 Editor assigned by journal 24 Apr, 2026 Editor invited by journal 20 Apr, 2026 Submission checks completed at journal 13 Apr, 2026 First submitted to journal 13 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9343223","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":632542899,"identity":"26c11b7f-695e-4944-aa2a-6d95efd1f9ff","order_by":0,"name":"Anne-Lise Jouen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA80lEQVRIiWNgGAWjYNCCAgkQmQAi5BhA7AqCWgwQWozBWs4Q1oJgJjYQ0sI/u/fgAwYDC3t+BoaHj3kq7qRvuN18gOHgHtxaJO6cSzYAOixxZgNDsjHPmWe5G+4cS2A48AyPNTdyzKT/GEgkGBxgSJPmbTucu+FGjgHzhwO4dcjfyDH/AbTF3h6s5d/hdIMb+R8YDuDRYgC0BRRijBsYQFoaDicARRjwajG8kWMsAfLLjMMMyYZzjh02nHnnmMEBfFrkbuQYfmCoqLPnb+9JfPCm5rA83+3mhw/waUEAZp4EOJsoDUDATqzCUTAKRsEoGGkAAIPLUwEI5GGiAAAAAElFTkSuQmCC","orcid":"","institution":"Université Bourgogne Europe, INSERM, CAPS UMR 1093","correspondingAuthor":true,"prefix":"","firstName":"Anne-Lise","middleName":"","lastName":"Jouen","suffix":""},{"id":632542900,"identity":"ff51f8f2-26d5-43e2-a28c-d29ea3078614","order_by":1,"name":"Ahmad Kaddour","email":"","orcid":"","institution":"Université Bourgogne Europe, INSERM, CAPS UMR 1093","correspondingAuthor":false,"prefix":"","firstName":"Ahmad","middleName":"","lastName":"Kaddour","suffix":""},{"id":632542901,"identity":"537f96d6-b498-4883-ba4f-dc09d69fedc2","order_by":2,"name":"Florent Lebon","email":"","orcid":"","institution":"Universite Lyon 1, LIBM, UR 7424","correspondingAuthor":false,"prefix":"","firstName":"Florent","middleName":"","lastName":"Lebon","suffix":""},{"id":632542902,"identity":"c13723a1-c290-4c67-b2a3-5a0aea1ef2c4","order_by":3,"name":"Peter Ford Dominey","email":"","orcid":"","institution":"Université Bourgogne Europe, INSERM, CAPS UMR 1093","correspondingAuthor":false,"prefix":"","firstName":"Peter","middleName":"Ford","lastName":"Dominey","suffix":""},{"id":632542903,"identity":"25f1e8a2-f9b6-4147-bed9-bcf49268ddc0","order_by":4,"name":"Jocelyne Ventre-Dominey","email":"","orcid":"","institution":"Université Bourgogne Europe, INSERM, CAPS UMR 1093","correspondingAuthor":false,"prefix":"","firstName":"Jocelyne","middleName":"","lastName":"Ventre-Dominey","suffix":""}],"badges":[],"createdAt":"2026-04-07 10:10:58","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9343223/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9343223/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108944285,"identity":"fbc51573-1260-425f-b323-c5073623a4c6","added_by":"auto","created_at":"2026-05-11 05:58:12","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":303670,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRaindcloud plots of the Forearm Bisection measures before (Pre) and After (Post) the robotic pointing task, in Experiment 1 - AO (left) and Experiment 2 - AOMI (right).\u003c/strong\u003e \u003cbr\u003e\nFor each experiment, the Raincloud Plot displays Pre vs Post observations (dots) and boxplots (median line; hinges = 25th/75th percentiles; whiskers = extremes within 1.5 × IQR), as well as a representation of smoothed density. Values are expressed in meters (cm).\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9343223/v1/d6b217619572c76d9bd3fc89.png"},{"id":108944313,"identity":"9dd86ef1-38fe-4e68-8095-e39b672b2566","added_by":"auto","created_at":"2026-05-11 05:58:25","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":304914,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRaindcloud plots of the Reaching distance estimation measures before (Pre) and After (Post) the robotic pointing task, in Experiment 1 - AO (left) and Experiment 2 - AOMI (right)\u003c/strong\u003e. \u003cbr\u003e\n For each experiment, the Raincloud Plot displays Pre vs Post observations (dots) and boxplots (median line; hinges = 25th/75th percentiles; whiskers = extremes within 1.5 × IQR), as well as a representation of smoothed density. Values are expressed in meters (m).\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9343223/v1/2e1ce48970d0275b784207db.png"},{"id":108944322,"identity":"14898bee-7daa-463a-898d-5c73013327f6","added_by":"auto","created_at":"2026-05-11 05:58:27","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":114123,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEmbodiment scores after the telepresence session in Experiment 1 - AO and Experiment 2 - AOMI. \u003c/strong\u003eMean embodiment scores in the AO (white lines) and AOMI (black lines) experiments for Ownership (OWN), Location (LOC), Agency (AG), and Control (C). Error bars indicate ± SD.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9343223/v1/337c9935a3c9b0877ab271bd.png"},{"id":108944283,"identity":"ce0b1d41-0aae-4a58-a623-10f6d1252afc","added_by":"auto","created_at":"2026-05-11 05:58:12","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":522064,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePhotographs of the humanoid robot iCub and experimental setup used in the telepresence experiments.\u003c/strong\u003e\u003cbr\u003e\n(a) Robotic telepresence setup: the participant wears a HMD and receives real-time visual input from the robot’s cameras. (b) Reaching distance estimation task used to assess PPS, with a red ball moving along a rail toward or away from the robot. (c) Robotic pointing task in which the robot points toward coloured cubes arranged on a grid. (d) Same robotic pointing task viewed from the robot first-person perspective as seen by the participant through the HMD during telepresence\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9343223/v1/c26ab0c9868278897fc98b07.png"},{"id":108944357,"identity":"25daae01-280c-4834-905c-e58805d5f6f2","added_by":"auto","created_at":"2026-05-11 05:58:40","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":445592,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eParallel experimental designs of the Experiment 1 - AO and Experiment 2 - AOMI protocols.\u003c/strong\u003e \u003cbr\u003e\nThe two experiments shared an identical structure and behavioural measurements, with the only difference being the inclusion of motor imagery evaluation/training and in-task ratings in Experiment 2.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-9343223/v1/5eedfb55ef739fbf9589ec4a.png"},{"id":108944454,"identity":"453d0305-f57a-4c57-a8d0-75d734d40fb9","added_by":"auto","created_at":"2026-05-11 05:59:06","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2050780,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9343223/v1/fe3cd5b4-5905-4725-a7ac-5a976e333cd0.pdf"},{"id":108944311,"identity":"3248e1d7-47db-481f-b379-3f918e10ad8d","added_by":"auto","created_at":"2026-05-11 05:58:24","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":14742,"visible":true,"origin":"","legend":"","description":"","filename":"SUPPLEMENTARYMATERIALS.docx","url":"https://assets-eu.researchsquare.com/files/rs-9343223/v1/a0ab7bd7f65c40159baa02cb.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Embodied Motor Imagery reshapes body and space representations","fulltext":[{"header":"Introduction","content":"\u003cp\u003eHumans have the remarkable ability to embody external objects, tools, or even artificial agents, dynamically incorporating their visuo-spatial features into body schema and peripersonal space (PPS) representations. This phenomenon, commonly referred to as embodiment, relies on the integration of multisensory and motor-related processes that support the sense of owning a body, localizing oneself in space, and acting as an agent in the environment [1\u0026ndash;3]. Such a visuo-spatial embodiment is supported by brain plasticity allowing the update of sensorimotor maps in response to environmental context and visuo-motor interactions. Both the body schema and PPS constitute core components of embodiment theory, as they form the functional interface between sensory experience and bodily action. Rather than being static representations, they dynamically structure the relationship between the self and the surrounding environment, notably by organizing visual space according to action possibilities [4].\u003c/p\u003e \u003cp\u003eAccordingly, embodiment entails a functional organization of visual space relative to the body and its action capabilities. Within this framework, visual space is typically distinguished into two main components: a near egocentric space (peripersonal), corresponding to hand-reaching actions, and a far retinocentric space (extrapersonal), extending beyond immediate reach and often associated with social interaction [5,6]. Supported by this dual organization, spatial encoding is flexible and context-dependent. For instance, studies in both animals and humans have shown that the boundaries of PPS can expand toward far space when using tools to reach distant objects [7\u0026ndash;10]. In parallel, tool use can reshape the body schema itself by altering the perceived metrics of the body [11,12]. Such sensory\u0026ndash;motor manipulations of body and space representations can also emerge during interactions with artificial or robotic effectors. For instance, synchronous visuo-motor movements with a detached robotic hand have been shown to shift the perceived location of one\u0026rsquo;s own hand toward the robot, even in the absence of tactile feedback or strong ownership sensations [13], suggesting that non-biological effectors can be partially incorporated into the body schema.\u003c/p\u003e \u003cp\u003eIn this context, robotic telepresence (i.e. setup in which a user remotely perceives and interacts with the environment through the sensors and effectors of a robotic surrogate) offers a unique framework to explore embodiment at a broader scale, defined as the conscious experience of the self as an acting body[14]. By providing a first-person visual perspective via head-mounted displays (HMD) and synchronized visual-motor feedback, robotic telepresence enables users to experience incorporation into a robot body. Previous work has shown that first-person visual and multisensory feedback can already induce whole-body embodiment of a walking humanoid robot surrogate, even with partial or delayed control of the robot\u0026rsquo;s movements[15]. This \"beaming\" into the robot can lead to a sense of ownership, agency, and self-location within the robot, even if its morphology differs from the human body[16]. Crucially, such embodied telepresence has been shown to increase robot acceptability and social closeness[17,18]. These effects suggest that embodiment is not only a perceptual illusion but also carries broader social and cognitive consequences.\u003c/p\u003e \u003cp\u003eIn the neuroscience perspective, embodiment is theorized to emerge from the integration of multisensory cues -including vision, proprioception, and touch- for the experience of ownership and location, along with intentional motor signals for the sense of agency. While passive visuo-motor synchrony can induce basic ownership illusions, as seen in the classic rubber hand illusion[19], the experience of agency requires internally generated motor commands, even if covert [20,21].\u003c/p\u003e \u003cp\u003eThese principles are formalized in motor simulation theory[22,23], which posits that action-related neural processes can be internally activated in the absence of actual movement execution. Within this framework, Action Observation (AO) and Motor Imagery (MI) are understood as two complementary forms of covert motor simulation, capable of engaging motor networks without overt movement. AO operates as a bottom-up process, triggered by the observation of others' movements. It recruits the mirror neuron system, originally described in primates [24,25], and supports embodied functions such as imitation, intention understanding, and action prediction. AO is a widely recognized approach for enhancing motor skill instruction and accelerating the learning process [26,27]. Its effectiveness depends on factors such as the observer's attention to task-relevant features and the contextual relevance of the observed action [28,29]. In contrast, MI is a top-down process in which individuals consciously simulate an action, generating visual and kinesthetic representations without producing any overt movement[30]. During MI, individuals consciously simulate movements, triggering predictive models of their expected sensory consequences [31\u0026ndash;33]. This simulation can support stronger embodiment by aligning predicted and observed outcomes. Importantly, MI and action execution recruit overlapping fronto-parietal circuits that are central to body schema and motor representations [22,34]. Because MI provides direct access to predicted sensory consequences, it is often regarded as the prototypical form of motor simulation and has been shown to enhance motor performance while promoting plastic changes within motor and body-related representations[35,36].\u003c/p\u003e \u003cp\u003eRecent studies highlight the particular effectiveness of combined AO and MI (AOMI), in which individuals simultaneously observe and mentally simulate the same action[37,38], in immersive and embodied settings, predominantly investigated in virtual reality contexts[39\u0026ndash;41]. Virtual reality and robotic interfaces effectively provide first-person perspectives, realistic avatars, and synchronized feedback that enhance ownership, agency, and neural engagement [42\u0026ndash;44]. Such designs have demonstrated improved motor performance, learning, and rehabilitation outcomes [45\u0026ndash;47]. Notably, the efficacy of these immersive protocols appears to depend on the richness of sensory feedback, the degree of immersion, and the integration of intentional motor simulation. In the context of robotic embodiment, several studies have explored the use of MI within brain-computer interface (BCI) systems. For instance, Alimardani et al., [44] showed that imagining movements to control a humanlike android robot could elicit a strong sense of agency and embodiment in participants, even without physical execution. They further demonstrated that performance-related feedback enhanced both the subjective sense of embodiment and the effectiveness of MI in a short time, suggesting a reciprocal relationship between MI skill and embodiment. Similarly, Ono et al.,[45] developed a BCI-controlled robotic exoskeleton that provided proprioceptive feedback during hand-squeezing tasks, showing promising results for motor rehabilitation. These approaches underscore the potential of combining motor simulation and embodied robotic interaction to strengthen sensorimotor representations.\u003c/p\u003e \u003cp\u003eWhile previous robotic studies have focused on active MI combined with BCI control and neurofeedback to induce embodiment[44,45], these paradigms involve deliberate interaction with the robotic effector. In contrast, our approach relies on passive telepresence based on AO or AOMI. We test whether internally generated motor simulation alone can modulate embodiment and body\u0026ndash;space representations, even in the absence of explicit control or proprioceptive feedback. Indeed, despite these advances, the specific role of intentional motor signals in modulating body schema and PPS during robotic telepresence remains underexplored. While observation alone may recalibrate PPS by updating reachability estimates, reshaping the body schema to match a robotic effector may require intentional motor signals to remap limb and space metrics in an embodied model. Likewise, previous research suggests that, depending on the intentional content of the motor signal, agency over a virtual robotic limb can contract or expand PPS and body schema representations[48].\u003c/p\u003e \u003cp\u003eIn the current study, we directly compare these mechanisms using two distinct experiments conducted in the same embodied telepresence setup: one based on Action Observation only (AO experiment), and a second based on combined Action Observation and Motor Imagery (AOMI experiment).\u003c/p\u003e \u003cp\u003eCritically, this design enables the investigation of embodied MI, in which participants imagine actions while being physically embodied in a real, moving robotic body viewed from a first-person perspective. Unlike most AOMI protocols typically conducted offline or in virtual environments, MI here unfolds within the physical world, with real-time visual, spatial, and environmental feedback (i.e., seeing, hearing and interacting with the experimenter while embodied in the robot). To our knowledge, such a direct comparison between AO and AOMI within the same fully embodied telepresence context remains largely unexplored.\u003c/p\u003e \u003cp\u003eWe hypothesize that: i) both conditions will modulate PPS and body representations, reflecting recalibration driven by the robot\u0026rsquo;s visual perspective and movements; ii) AOMI will lead to differential effects on body schema metrics, with changes reflecting the contribution of intentional motor processes; iii) Embodiment experiences (including ownership, location, and agency) will be modulated by the task, with AOMI expected to promote stronger or more integrated embodiment compared to AO alone.\u003c/p\u003e \u003cp\u003eBy clarifying the contributions of these mechanisms, our study aims to advance the understanding of how motor simulation and 1st person perspective contribute to embodiment and plasticity of spatial/body representation during robotic telepresence.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003eExperiment 1 \u0026ndash; Action Observation (AO)\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eBehavioral measurements\u003c/h2\u003e \u003cp\u003eDuring the robotic pointing task in the AO condition (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), the mean estimated position of the forearm bisection (BIS; behavioral measure of perceived forearm midpoint) was 19.50 cm (SD\u0026thinsp;=\u0026thinsp;4.77) at pre-test and 18.48 cm (SD\u0026thinsp;=\u0026thinsp;4.12) at post-test. This difference was not statistically significant (main effect of Time: F(1, 20)\u0026thinsp;=\u0026thinsp;2.37, p = .14, η\u0026sup2;p = .086), indicating no reliable change in perceived forearm midpoint following the robotic pointing task.\u003c/p\u003e \u003cp\u003eIn contrast, PPS (RD - reaching distance estimation, behavioral measure of perceived reachability boundary) changed significantly over time (main effect of Time: F(1, 20)\u0026thinsp;=\u0026thinsp;13.09, p = .002, η\u0026sup2;p = .38). As illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, the estimated PPS boundary decreased from pre-test (M\u0026thinsp;=\u0026thinsp;0.61, SD\u0026thinsp;=\u0026thinsp;0.09) to post-test (M\u0026thinsp;=\u0026thinsp;0.57, SD\u0026thinsp;=\u0026thinsp;0.10).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eEmbodiment questionnaire\u003c/h3\u003e\n\u003cp\u003eFor each of the three embodiment categories (OWN - Ownership, LOC \u0026ndash; Location, AG - Agency; questionnaire-based measures of embodiment), the sensation of embodiment into the robot was significantly different from 0 (all p \u0026lt; .001). The ANOVA also revealed a significant main effect of Category (F(3, 66)\u0026thinsp;=\u0026thinsp;6.13, p \u0026lt; .001, η\u0026sup2;p = .22). Post-hoc tests showed that only OWN (p = .0022) and LOC (p = .0036) differed significantly from the Control (CTRL) category (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFor each experiment, the Raincloud Plot displays Pre vs Post observations (dots) and boxplots (median line; hinges\u0026thinsp;=\u0026thinsp;25th/75th percentiles; whiskers\u0026thinsp;=\u0026thinsp;extremes within 1.5 \u0026times; IQR), as well as a representation of smoothed density. Values are expressed in meters (cm).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFor each experiment, the Raincloud Plot displays Pre vs Post observations (dots) and boxplots (median line; hinges\u0026thinsp;=\u0026thinsp;25th/75th percentiles; whiskers\u0026thinsp;=\u0026thinsp;extremes within 1.5 \u0026times; IQR), as well as a representation of smoothed density. Values are expressed in meters (m).\u003c/p\u003e\n\u003ch3\u003eDiscussion Experience 1 - AO\u003c/h3\u003e\n\u003cp\u003eThe results of Experiment 1 show that passive AO during embodied robotic telepresence is sufficient to induce selective modifications of spatial representations, as revealed by a significant contraction of PPS following the robotic pointing task. This effect likely reflects an updating of reachability estimates driven by the robot\u0026rsquo;s visual perspective and limb metrics. In contrast, no reliable change was observed in body schema as indexed by the forearm bisection task, suggesting that visual observation alone does not suffice to rescale internal representations of the body.\u003c/p\u003e \u003cp\u003eAt the subjective level, participants reported significant sensations of ownership and self-location within the robot, while agency remained comparatively weak. Together, these findings indicate that first-person visual exposure to a robotic body can support partial embodiment and spatial recalibration, but that the absence of intentional motor engagement limits both the emergence of agency and the plasticity of body schema representations.\u003c/p\u003e \u003cp\u003eTaken together, these findings suggest that passive observation alone may be insufficient to modify internal body schema representations. We therefore asked whether introducing intentional motor simulation and enhancing agency through MI would produce additional changes. Experiment 2 was designed to address this question using an embodied AOMI condition.\u003c/p\u003e \u003cp\u003eExperiment 2 \u0026ndash; Action Observation\u0026thinsp;+\u0026thinsp;Motor Imagery (AOMI)\u003c/p\u003e\n\u003ch3\u003eBehavioral measurements\u003c/h3\u003e\n\u003cp\u003eDuring the robotic pointing task in the AOMI condition (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), the estimated forearm bisection (BIS; behavioral measure of perceived forearm midpoint) shifted significantly toward the trunk from pre-test (M\u0026thinsp;=\u0026thinsp;22.37 cm, SD\u0026thinsp;=\u0026thinsp;3.63) to post-test (M\u0026thinsp;=\u0026thinsp;19.83 cm, SD\u0026thinsp;=\u0026thinsp;3.08). This change was statistically significant (main effect of Time: F(1, 20)\u0026thinsp;=\u0026thinsp;11.27, p = .003, η\u0026sup2;p = .36), indicating a contraction of the perceived forearm metric representation following the robotic pointing task.\u003c/p\u003e \u003cp\u003ePPS (RD - reaching distance estimation, behavioral measure of perceived reachability boundary) also changed significantly over time (main effect of Time: F(1, 20)\u0026thinsp;=\u0026thinsp;16.47, p \u0026lt; .001, η\u0026sup2;p = .45). As illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, the estimated PPS boundary decreased from pre-test (M\u0026thinsp;=\u0026thinsp;0.60, SD\u0026thinsp;=\u0026thinsp;0.04) to post-test (M\u0026thinsp;=\u0026thinsp;0.56, SD\u0026thinsp;=\u0026thinsp;0.05).\u003c/p\u003e\n\u003ch3\u003eEmbodiment questionnaire\u003c/h3\u003e\n\u003cp\u003eThe AOMI experiment induced a significant sensation of embodiment, with scores significantly different from zero across categories (p \u0026lt; .001). The ANOVA revealed a significant main effect of Category (F(3, 60)\u0026thinsp;=\u0026thinsp;19.42, p \u0026lt; .001, η\u0026sup2;p = .49). Post-hoc tests showed that OWN (M\u0026thinsp;=\u0026thinsp;37, SD\u0026thinsp;=\u0026thinsp;18), LOC (M\u0026thinsp;=\u0026thinsp;59, SD\u0026thinsp;=\u0026thinsp;26), and AG (M\u0026thinsp;=\u0026thinsp;41, SD\u0026thinsp;=\u0026thinsp;25) were all significantly higher than the CTRL category (M\u0026thinsp;=\u0026thinsp;0.14, SD\u0026thinsp;=\u0026thinsp;15; p \u0026lt; .01). Additionally, LOC scores were higher than OWN (p = .002) and AG (p \u0026lt; .02).\u003c/p\u003e \u003cp\u003eAs shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, location and agency values were substantially higher in the AOMI condition than in AO alone.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eCorrelations Between Motor Imagery and Embodiment questionnaires measures\u003c/h2\u003e \u003cp\u003eCorrelation analyses examined the relationship between MI vividness ratings collected during the robotic pointing task and embodiment subscales (OWN, LOC, AG, and CTRL). Four Pearson correlations were computed. Two uncorrected correlations reached significance: AG\u0026ndash;MI (r = .48, p = .031) and LOC\u0026ndash;MI (r = .56, p = .010). Because multiple tests were performed, p-values were adjusted using the Bonferroni correction (α_adj = .0125). After correction, only the LOC\u0026ndash;MI correlation remained significant (p = .010). Agency\u0026ndash;MI (p = .031), Ownership\u0026ndash;MI (r = .11, p = .630), and CTRL\u0026ndash;MI (r = .17, p = .475) did not reach the corrected threshold.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eDiscussion Experience 2 – AOMI\u003c/h3\u003e\n\u003cp\u003eExperiment 2 demonstrates that adding an explicit MI component during action observation in robotic telepresence qualitatively modifies the pattern of embodiment and its consequences on body representations. In contrast to AO alone, combined AOMI induced a significant shift in forearm midpoint estimates toward the trunk along with a reduction of PPS. This suggests that intentional motor simulation contributes to recalibrating both bodily and spatial metrics during embodied interaction with a robotic body.\u003c/p\u003e \u003cp\u003eSubjective embodiment was also stronger and more differentiated in the AOMI condition, with ownership, self-location, and agency all exceeding control levels, and location emerging as the dominant dimension. Importantly, the correlation between MI vividness and self-location indicates that the degree of internal motor engagement directly relates to how strongly participants locate themselves within the robot. Together, these findings support the view that embodied MI provides a critical mechanism for reshaping internal body representations during robotic telepresence, beyond the effects of visual observation alone.\u003c/p\u003e \u003cp\u003e \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eSynthesis of Experiments 1 - AO and 2 - AOMI: Effects of Observation and Motor Imagery on Embodiment and Body Representation\u003c/span\u003e \u003c/p\u003e \u003cp\u003eAs summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, the addition of MI to AO produced broader effects, encompassing body schema and agency, and underscoring a dissociation between PPS and body schema modulation.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEffects of action observation and motor imagery on peri-personal space, body schema, and embodiment\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e\u003cb\u003eAction Observation (Exp1)\u003c/b\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePPS (RD)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBody Schema (BIS)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eEmbodiment\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eReduced\u003c/p\u003e \u003cp\u003eη\u0026sup2;p = .38\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eUnchanged\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eOWN, LOC\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAction Observation and Motor Imagery (Exp2)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eReduced\u003c/p\u003e \u003cp\u003eη\u0026sup2;p = .45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReduced\u003c/p\u003e \u003cp\u003eη\u0026sup2;p = .36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eOWN, LOC, AG\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e"},{"header":"GENERAL DISCUSSION","content":"\u003cp\u003eIn this study, we used robotic telepresence to investigate how embodiment within a small-sized humanoid robot modulates spatial and bodily representations, specifically peripersonal space and body schema. Action observation (AO) alone induced a contraction of peripersonal space, whereas combining action observation with motor imagery (AOMI) additionally altered both peripersonal space and body schema. These findings indicate that embodiment-related plasticity depends not only on first-person visual perspective but also on intentional motor engagement. Thus, the nature of motor signals critically shapes how the body and surrounding space are internally represented during telepresence.\u003c/p\u003e \u003cp\u003eThese results can be interpreted in light of previous work demonstrating that tool use reshapes both peripersonal space and body schema, extending spatial representations and altering body metrics through sensorimotor interaction with the environment [7\u0026ndash;12]. Consistent with this literature, the contraction of peripersonal space observed during AO likely reflects recalibrated reachability estimates based on the robot\u0026rsquo;s shorter arm metrics, similarly to tool-use paradigms in which new effectors are incorporated into spatial representations even without overt movement execution.\u003c/p\u003e \u003cp\u003eIn the present study, simply observing the movements of a shorter robotic limb from a first-person perspective was sufficient to induce changes in peripersonal space. Importantly, however, in this context, the moving robot arm cannot be reduced to the passive properties of a static instrument. Although the iCub robot\u0026rsquo;s arm appeared metallic, it functioned as an active effector rather than a tool, and immersive first-person telepresence likely promoted its interpretation as a surrogate body part This embodied integration may explain why passive observation alone modulated peripersonal space in our paradigm, whereas previous studies directly comparing active and passive tool use have shown that merely observing tool manipulation, without motor execution, is insufficient to induce reliable changes in body or spatial representations[49].\u003c/p\u003e \u003cp\u003eIn contrast, adding MI during telepresence produced qualitatively stronger effects. Combined AO and MI induced not only peripersonal space contraction but also significant changes in body schema, reflected by a perceived shortening of the forearm segment. Thus, the AOMI condition did not simply amplify the observation effect; it introduced an additional functional component, suggesting that internally generated motor signals are necessary to update internal body metrics. In consequence, even in the absence of overt motor execution, the association of visual and intentional signals was sufficient to modify body representations, in addition to inducing space remapping. Mechanistically, this supports the idea that body schema remapping requires stronger sensorimotor involvement, particularly intentional motor signals[20,21,31,48]. During MI, internally generated motor commands help predict the robot\u0026rsquo;s movements and facilitate their integration into the body schema. Consistent with this interpretation, a similar dissociation has been reported in healthy aging, where reduced reliability of internal bodily signals leads to distortions of body representations while peripersonal space remains relatively preserved[50,51], further supporting the idea that body schema updating relies more strongly on internally generated sensorimotor signals than peripersonal space.\u003c/p\u003e \u003cp\u003eAt the subjective level, concurrent embodiment increased during AOMI compared with AO alone, particularly for the senses of self-location and agency. Similar embodiment phenomena have been reported in clinical contexts, such as post-stroke sensorimotor rehabilitation and in the integration and acceptance of prosthetic devices [52,53], where artificial limbs or surrogate body parts are progressively incorporated into the user\u0026rsquo;s body representation. This incorporation is thought to rely on sensorimotor mapping and visuo-motor congruence between real and artificial elements, allowing external effectors -including robotic limbs- to be integrated into internal body representations [16,17]. Together, our results indicate that visual immersion supports partial embodiment, whereas intentional motor engagement strengthens and deepens this incorporation. More broadly, our findings suggest that embodiment arises from both passive and active processes: passive visual observation can recalibrate peripersonal space and promote ownership and self-location, whereas the emergence of agency depends on intentional motor processing and the comparison between predicted and actual action outcomes[21,38].\u003c/p\u003e \u003cp\u003eThe enhanced embodiment observed during AOMI can be interpreted in light of theoretical models of motor simulation. The outcome of the AOMI situation is theorized to enhance motor simulation by coupling bottom-up visual input with top-down MI signals, leading to greater activation of sensorimotor networks than either process alone[54,55]. Two complementary models explain AOMI\u0026rsquo;s efficacy: the Dual Action Simulation hypothesis[37,38,56], which suggests parallel motor representations for observed and imagined actions, and the Visual Guidance Hypothesis[57], which proposes that the imagined action is prioritized while AO serves primarily as a visual guide that primes and stabilizes MI. In our telepresence setup, participants simultaneously observed and imagined the robotic movements, a configuration consistent with AOMI principles that likely amplified sensorimotor engagement and the sense of embodiment. Importantly, our results extend the literature findings by demonstrating that AOMI in a physically embodied robotic context does not merely improve motor-related processes but can directly reshape both spatial and bodily representations. Furthermore, the stronger embodiment observed during AOMI\u0026mdash;particularly the emergence of a sense of agency - and its correlation with imagery vividness suggests that motor simulation actively contributes to the construction of bodily self-representation. The greater sense of agency reported in the motor imagery condition is also consistent with previous AOMI research showing enhanced motor learning, stronger neural activation, and improved behavioral outcomes compared with action observation alone[54,55,58]. Moreover, immersive and embodied contexts such as virtual reality and robotics appear to further amplify these effects[41\u0026ndash;43].\u003c/p\u003e \u003cp\u003eIn terms of neural substrates, both MI and AO engage extensive motor-related brain networks, with substantial overlap between the two processes and motor execution. MI, in particular, involves widespread activation across the primary motor cortex (M1), supplementary motor area (SMA), premotor and parietal cortices, as well as prefrontal and subcortical structures[34,37,59\u0026ndash;61]. A meta-analysis of fMRI studies identified a shared core network\u0026mdash;including the premotor cortex, rostral parietal regions, and somatosensory areas\u0026mdash;commonly activated during MI, AO, and overt action[62]. Interestingly, MI more closely resembles physical execution network due to its stronger subcortical involvement. In contrast, AO relies more heavily on visual input, activating shared motor representations via the mirror neuron system and engaging premotor and parietal regions involved in action understanding[59\u0026ndash;61]. Importantly, studies have shown that combining AO and MI in mixed protocols enhances both neural activation and behavioral outcomes beyond what either technique achieves alone[37,54,55,58].\u003c/p\u003e \u003cp\u003eThese previous findings provide a plausible neural substrate to account for the stronger embodiment effects observed in our study and highlight the added value of embodied AOMI protocols. Indeed, such approaches not only enhance motor performance but also reshape body schema and peripersonal space representations during telepresence, with important implications for teleoperation, rehabilitation, and assistive robotics.\u003c/p\u003e \u003cp\u003eBeyond its neural underpinnings, the dissociation observed between body schema and peripersonal space in our study further highlights the role of intentional motor control in shaping embodiment. In the current study, only the combined sensory and motor signals triggered in the AOMI experiment induced transformations in both body schema and peripersonal space, accompanied by an additional sense of agency. To explain the link between motor processing and the modulation of body schema and peripersonal space, we suggest that higher-order intentional processes related to motor control are crucial for generating the sense of agency and the consequent changes in spatial and body representations. The sense of agency relies on the comparison between motor commands issued from intentional processes and the feedback signals informing the sensory consequences of an action. In the case of MI, although the motor command is attenuated to prevent overt execution, a copy of this motor signal is generated and compared with afferent signals arising from the observed visual scene[20,21,31].\u003c/p\u003e \u003cp\u003eLikewise, in our previous studies on human\u0026ndash;robot interactions [16\u0026ndash;18], we described how intentional motor commands triggered during active visuo-motor interactions were responsible for external agency and changes in social perception. During these robotic telepresence experiments, shared synchronous movements between the robot and the human may engage resonance mechanisms supported by the mirror neuron system, which are associated with several social functions, including theory of mind, empathy, and self-recognition[63,64]. Similar resonance mechanisms may also be elicited when one imagines generating a hand movement in synchrony with observing the robot\u0026rsquo;s movement.\u003c/p\u003e \u003cp\u003eBased on these observations, we suggest that intentional commands underlying these simultaneous imagined or real movements are crucial for producing changes in body and spatial representations, as well as social feelings, co-occurring with the sensation of embodiment into the robot. This interpretation is consistent with theories of the mirror neuron network, which propose that resonance mechanisms in the brain allow shared action representations to emerge during robotic telepresence.\u003c/p\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eOur findings demonstrate that robotic telepresence can flexibly modulate both spatial and bodily representations in humans. More importantly, they reveal that robotic embodiment is a graded phenomenon shaped by the level of motor engagement. While visual observation alone is sufficient to recalibrate action-related space, the addition of motor imagery qualitatively extends this plasticity by engaging internal motor signals that contribute to body metric recalibration and strengthen the sense of agency. In this sense, AOMI does not merely amplify the effects of observation, but introduces an additional layer of embodiment, linking spatial remapping, bodily self-representation, and intentional motor processing within a unified framework.\u003c/p\u003e \u003cp\u003eBeyond these experimental insights, embodied AOMI protocols in robotic telepresence hold strong basic and translational potential. By engaging motor networks without overt movement, they offer a powerful paradigm to investigate the neural foundations of bodily self-representation and space awareness in immersive real-world contexts. At the same time, such approaches provide a concrete framework for neurorehabilitation and assistive technologies, positioning embodiment within robotic or virtual avatars as a valuable bridge between fundamental neuroscience and therapeutic innovation.\u003c/p\u003e"},{"header":"MATERIAL AND METHODS","content":"\u003cp\u003eExperiment 1 \u0026ndash; Action Observation (AO)\u003c/p\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e- Participants\u003c/h2\u003e \u003cp\u003eA group of 21 healthy, right-handed participants (10 males, mean age\u0026thinsp;=\u0026thinsp;24.38 years, SD\u0026thinsp;=\u0026thinsp;6.5) participated in Experiment 1 \u0026ndash; AO. All participants had normal or corrected-to-normal vision and reported no history of drug use, neurological, or psychiatric disorders. Participants were na\u0026iuml;ve to the purpose of the study and provided written informed consent prior to participation. The experimental protocol was approved by the Ethics Committee for Research of Universit\u0026eacute; Bourgogne Europe (CER UBE; approval number: CERUBFC-2021-07-10-028) and conducted in accordance with the Declaration of Helsinki (1964). Individuals appearing in the figures provided informed consent for publication of identifying images in an online open-access format.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e- Apparatus and Setup\u003c/h2\u003e \u003cp\u003eThe AO experiment employed a robotic telepresence setup in which participants wore a head-mounted display (HMD) providing a real-time, first-person visual perspective from the robot\u0026rsquo;s viewpoint. The study was conducted using the humanoid robot iCub (Metta et al., 2010), which is approximately the size of a 3-year-old child and equipped with video cameras in its eyes, enabling it to explore and perceive its visual environment. The stereo video signal from the two cameras is transmitted to the HMD (SONY HMZ-3WT 3D Viewer) to generate for the participant a stereo immersion in the scene from the robot\u0026rsquo;s perspective, including vision of the robot\u0026rsquo;s moving arms and hands, to induce a sense of embodiment into the robot body.\u003c/p\u003e \u003cp\u003eParticipants sat comfortably at a table in a quiet room, wearing the HMD connected to the robot\u0026rsquo;s camera system. The robot was placed elsewhere in the experimental room, out of the participant's view from the robot's perspective as illustrated in (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e-a). During the telepresence tasks, the robot executed a series of preprogrammed pointing movements with its hand and forearm. The dimensions of the robotic limb were smaller than those of a typical human limb, but this mismatch was not explicitly mentioned by the experimenter.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e(a) Robotic telepresence setup: the participant wears a HMD and receives real-time visual input from the robot\u0026rsquo;s cameras. (b) Reaching distance estimation task used to assess PPS, with a red ball moving along a rail toward or away from the robot. (c) Robotic pointing task in which the robot points toward coloured cubes arranged on a grid. (d) Same robotic pointing task viewed from the robot first-person perspective as seen by the participant through the HMD during telepresence\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e- Procedure\u003c/h2\u003e \u003cp\u003eAfter a general explanation of the tasks, the participant was familiarized with the iCub robot through a brief demonstration of the robot\u0026rsquo;s head and arm motor capabilities.\u003c/p\u003e \u003cp\u003eThe experimental procedure comprised five successive phases and lasted approximately 40 minutes in total (see Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The forearm bisection (BIS) and reaching distance (RD) were used as behavioral measurements of body schema and PPS representations, respectively, whereas the robotic pointing task constituted the main experimental task. All phases but the BIS measurement took place with the HMD i.e., during telepresence.\u003c/p\u003e \u003cp\u003eParticipants first completed a BIS measurement without the HMD. Then, they wore the HMD and performed a RD measurement, followed by the robotic pointing task, during which they passively observed the robot\u0026rsquo;s movements from a first-person perspective (AO experiment). Still wearing the HMD, participants subsequently performed a second RD measurement to assess changes in PPS. Finally, the HMD was removed and a final BIS measurement was conducted to evaluate changes in body representation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe two experiments shared an identical structure and behavioural measurements, with the only difference being the inclusion of motor imagery evaluation/training and in-task ratings in Experiment 2.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e- Behavioral measurements\u003c/h2\u003e \u003cdiv id=\"Sec17\" class=\"Section3\"\u003e \u003ch2\u003eo Forearm bisection (BIS)\u003c/h2\u003e \u003cp\u003eThe BIS measure was used to assess changes in body schema, specifically the perceived metric representation of the arm.\u003c/p\u003e \u003cp\u003eParticipants\u0026rsquo; right arm was positioned parallel to the midsagittal plane on the table and covered with a rigid screen to prevent visual and tactile feedback. While blindfolded, participants pointed with their left index finger to the perceived midpoint of their right forearm, defined as the segment between the elbow and the tip of the middle finger[48,65]. To minimize spatial prediction, the left hand was placed in different initial positions across trials. After three practice trials, participants completed 12 experimental trials with randomized starting positions. For each trial, the indicated midpoint was recorded using a graduated scale fixed to the rigid cover.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eo Reaching distance estimation (RD)\u003c/h2\u003e \u003cp\u003eThe RD measure, adapted from D\u0026rsquo;Angelo et al.[48], was used to assess the boundaries of PPS.\u003c/p\u003e \u003cp\u003eFrom the robot\u0026rsquo;s first-person perspective, viewed through the HMD, participants observed a table equipped with a 150-cm drylin\u0026reg; rail along which a red ball (8-cm diameter), mounted on a small platform, could move either toward (approaching trials) or away from (withdrawing trials) the robot (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e-b). The ball\u0026rsquo;s starting position (25 and 100 cm for withdrawing and approaching trials, respectively) and speed (approximately 2.75 cm/s) were precisely controlled by the experimenter using a joystick. During each trial, participants verbally indicated, without moving, when the ball reached the boundary at which it was judged reachable (approaching trials) or unreachable (withdrawing trials), following the instruction: \u0026ldquo;Say stop when you estimate that you could or could not reach the ball.\u0026rdquo; The final stopping position was recorded using a laser telemeter measuring the distance between the robot\u0026rsquo;s midsagittal axis and the ball location at the stop position.\u003c/p\u003e \u003cp\u003eBetween two trials, participants briefly closed their eyes while the experimenter repositioned the ball for the next trial. After two practice trials, participants completed a total of 12 experimental trials, including six trials for each motion direction (toward and away from the robot).\u003c/p\u003e \u003cp\u003eBoth the BIS and RD measurements were administered twice, once before and once after the AO pointing task, using the same protocol.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e- Robotic pointing task\u003c/h2\u003e \u003cp\u003eIn the robotic pointing task, participants wearing an HMD viewed the robot\u0026rsquo;s forearm from a first-person perspective as it pointed toward colored cubes arranged on a chessboard (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e-c). A 75 \u0026times; 60 cm tilted grid tablet was positioned in front of the robot and centered on its body axis. Prior to the experiment, the robot\u0026rsquo;s PPS boundary was objectively determined through preliminary reaching measurements to estimate the maximum reachable workspace of the robotic arm. This calibration was used to spatially classify targets as reachable (PPS) or unreachable (extrapersonal space EPS), resulting in an intentionally asymmetric distribution of cubes (5 within PPS and 7 beyond reach, in EPS) due to the smaller size of the robot\u0026rsquo;s reachable workspace. On the tablet, the 12 cubes (5 cm\u0026sup3;) were arranged to cover the grid accordingly.\u003c/p\u003e \u003cp\u003eAt each trial, the robot pointed with its right hand to a cube that had been previously selected and lifted for 2 seconds by the experimenter as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e-c, d. While the robot hand was pointing to a cube, the experimenter verbally indicated whether the robot hand was touching the corresponding cube or not, and then returned the cube to its initial position. To optimize coordinated reaching movements and visual fixation, the robot\u0026rsquo;s motion was programmed using a combination of trunk rotation, forearm displacement, and head orientation (iCub \u003cem\u003ectpservice\u003c/em\u003e function) to ensure maximal accuracy across spatial positions. The sequences of pointing movements were pre-established, and the experimenter was cued before each trial regarding the target location. A video demonstrating this behavior can be seen: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://youtu.be/aYQlfk9EOVw\u003c/span\u003e\u003cspan address=\"https://youtu.be/aYQlfk9EOVw\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eThe pointing task consisted of 10 sequences of 10 pointing movements executed by the robot toward cubes located either within or beyond reaching distance. The order of cubes and sequences was randomized across participants to ensure a balanced number of reachable and non-reachable trials. Through the HMD, participants could clearly perceive the movements of the robot\u0026rsquo;s hand, fingers, and forearm (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e-d).\u003c/p\u003e \u003cp\u003eDuring the robotic pointing task, participants were instructed to carefully observe the spatial positions of the cubes, the robot\u0026rsquo;s hand and forearm movements, and whether each target was within or beyond the robot\u0026rsquo;s reachable space (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e-d). They were asked to remain relaxed, refrain from any physical movement, and simply observe the robot\u0026rsquo;s pointing gestures from a first-person perspective through the HMD. No explicit instructions regarding embodiment or perspective-taking were provided, allowing any sensations of embodiment to emerge naturally. The session consisted of 10 sequences of 10 pointing movements (100 movements in total) and lasted approximately 10 minutes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e- Embodiment questionnaire\u003c/h2\u003e \u003cp\u003eAt the end of the experiment, participants were invited to complete a questionnaire aimed at quantifying their sensation of embodiment into the robot during the robotic pointing task. This questionnaire, consisting of 10 statements (see Supplementary Materials: SM1), allowed evaluation of embodiment on a subjective scale ranging from 0 (no sensation) to 100 (extremely strong sensation). For each statement, participants indicated a value between 0 and 100 specifying the level of embodiment felt during the experiment.\u003c/p\u003e \u003cp\u003eThe embodiment score was assessed across three categories of statements focusing on: (i) the sensation that the robot\u0026rsquo;s hand belonged to the participant (3 items), corresponding to ownership (OWN); (ii) the sensation that the participant\u0026rsquo;s hand was located at the position of the robot\u0026rsquo;s hand, or that they were located at the robot\u0026rsquo;s position (2 items), corresponding to self-location (LOC); and (iii) the sensation of controlling the robot\u0026rsquo;s movements (3 items), corresponding to agency (AG), reflecting the perceived motor control exerted over the robot. Two control statements, assumed to be independent of the experimental manipulation, were also included. Scores for each category were computed as the average of the ratings across the corresponding items.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eData analysis\u003c/h2\u003e \u003cp\u003eStatistical analyses and figures were performed using Statistica (TIBCO Software Inc., CA 94304, USA) and JASP[66]. Data were first assessed for normality (Kolmogorov\u0026ndash;Smirnov) and homogeneity of variances (Levene\u0026rsquo;s test). As all assumptions were met (p \u0026gt; .05), parametric tests were applied. The significance level was set at α\u0026thinsp;=\u0026thinsp;.05. Mixed repeated-measures ANOVAs were conducted for BIS (i.e., estimated midpoint of the hand\u0026ndash;forearm segment) and RD (estimation of the PPS boundary) measures, with Time (Pre, Post) as a within-subject factor. Embodiment questionnaire scores were analyzed using repeated-measures ANOVAs with Category (Ownership OWN, Location LOC, Agency AG, Control CTRL) as a within-subject factor. Significant effects were followed by Bonferroni-corrected pairwise comparison.\u003c/p\u003e \u003cp\u003eExperiment 2 \u0026ndash; Action Observation\u0026thinsp;+\u0026thinsp;Motor Imagery (AOMI)\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e- Participants\u003c/h2\u003e \u003cp\u003eA different group of 21 participants (8 males, mean age\u0026thinsp;=\u0026thinsp;22.89 years, SD\u0026thinsp;=\u0026thinsp;5.0) participated in Experiment 2 \u0026ndash; AOMI. Inclusion criteria were identical to those described for Experiment 1 \u0026ndash; AO. All participants provided written informed consent prior to participation. The study was approved by the Ethics Committee for Research of Universit\u0026eacute; Bourgogne Europe (CER UBE; approval number: CERUBFC-2021-07-10-028) and conducted in accordance with the Declaration of Helsinki.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003e- Apparatus and Setup\u003c/h2\u003e \u003cp\u003eThe same robotic telepresence apparatus, HMD, and iCub humanoid robot described in Experiment 1 were used.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003e- Procedure\u003c/h2\u003e \u003cp\u003eThe overall procedure was identical to that of Experiment 1, including the sequence and timing of the behavioral measurements (BIS and RD) and the robotic pointing task (see Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The only differences concerned the instructions given to the participants and the addition of a MI training and evaluation phase prior to telepresence exposure, as well as intermittent MI vividness ratings collected between blocks during the robotic pointing task. The total duration of the Experiment 2 - AOMI was approximately 45\u0026ndash;50 minutes.\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003e- MI evaluation and training\u003c/h2\u003e \u003cp\u003ePrior to the behavioral measurements, participants\u0026rsquo; intrinsic MI vividness was assessed using the KVIQ-10, a short version of the Kinesthetic and Visual Imagery Questionnaire (KVIQ-20; [67]). The KVIQ evaluates, on a five-point ordinal scale, the clarity of visual imagery and the intensity of kinesthetic sensations experienced during first-person MI. Participants were seated comfortably and performed 10 simple movements, each followed by the corresponding imagined movement (performed with eyes open), which they rated using both visual and kinesthetic subscales.\u003c/p\u003e \u003cp\u003eFor MI training, participants sat in front of a table on which three colored cubes were positioned approximately 40 cm away, reproducing the spatial configuration and viewing angle of the robotic pointing task. First, participants physically performed pointing movements toward each cube (four repetitions per cube, 12 trials total). Second, they were asked to imagine performing the same pointing gestures while focusing on both visual and kinesthetic sensations. After each imagined trial, participants rated the vividness of their imagery on a scale from 0 (no sensation) to 5 (as if actually moving), reflecting both kinesthetic and visual aspects.\u003c/p\u003e \u003cp\u003eAfter this MI training phase, participants proceeded with the first BIS and RD measurements, as described in Experiment 1 - AO.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003e- Behavioral measurements and embodiment questionnaire\u003c/h2\u003e \u003cp\u003eThe BIS, RD, and embodiment questionnaire measures were administered and processed in the same manner as in Experiment 1 \u0026ndash; AO.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003e- Robotic pointing task\u003c/h2\u003e \u003cp\u003eParticipants then performed the robotic pointing task. Using the same setup, they observed identical robotic pointing movements through the HMD for approximately 10 minutes, as in the AO experiment. The key difference in this condition was that they were explicitly instructed to engage in MI, imagining themselves performing the same pointing movements in real time as precisely as possible. During the task, they observed the robot\u0026rsquo;s hand and forearm, the spatial location of each cube, and whether it was reachable or not by the robot, while simultaneously imagining executing the same movement. They were asked to focus on the kinesthetic sensations, mentally simulating the feeling of their own arm and hand moving in synchrony with the robot. Each sequence included 10 robotic pointing movements and was repeated until all 10 movements were completed, for a total of 100 movements. At the end of each 10-movement sequence, participants rated their imagery vividness on a scale from 0 (no sensation at all) to 5 (as if actually moving), reflecting both kinesthetic and visual impressions, as in the earlier MI training.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003eData analysis\u003c/h2\u003e \u003cp\u003eStatistical analyses followed the same procedure as described for Experiment 1 \u0026ndash; AO.\u003c/p\u003e \u003cp\u003eAdditional analyses were conducted to examine the MI component. KVIQ scores and imagery vividness ratings obtained during training indicated that participants had an average KVIQ total score of 30.90/50 and an average training vividness score of 3.22/5; these measures were therefore not further analyzed. MI vividness ratings collected during the robotic pointing task were used to verify that participants maintained a consistent level of imagery vividness throughout the session, thereby limiting potential mental fatigue[68]. These MI ratings were also correlated with embodiment questionnaire scores using Pearson\u0026rsquo;s correlation coefficient (r).\u003c/p"},{"header":"Declarations","content":"\u003cul type=\"disc\"\u003e\n \u003cli\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eJ.V.D. and. P.F.D conceived the original research idea. All authors contributed to the study conception and experimental design. A.K. and P.F.D. developed and implemented the robotic platform.\u0026nbsp;\u003cbr\u003eA.L.J. conducted participant recruitment, and A.L.J. and J.V.D. carried out data collection and analyses. A.L.J. and J.V.D. drafted the initial manuscript. All authors contributed to the interpretation of the results and provided critical revisions of the manuscript. A.L.J. prepared the final version of the manuscript. All authors reviewed and approved the final manuscript and agreed to its submission to \u003cem\u003eScientific Reports\u003c/em\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003e\u003cstrong\u003eData availability statement\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThe datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003e\u003cstrong\u003eAdditional Information (including a Competing Interests Statement)\u003c/strong\u003e\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003e\u003cstrong\u003eFunding Declaration\u003c/strong\u003e\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThis research was supported by the French Region Bourgogne-Franche-Comt\u0026eacute; (Grant ANER RobotSelf 2019-Y-10650).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eGallagher, S. Dimensions of Embodiment: Body Image and Body Schema in Medical Contexts. in \u003cem\u003eHandbook of Phenomenology and Medicine\u003c/em\u003e (ed. Toombs, S. K.) 147\u0026ndash;175 (Springer Netherlands, Dordrecht, 2001). doi:10.1007/978-94-010-0536-4_8.\u003c/li\u003e\n\u003cli\u003ede Vignemont, F. Embodiment, ownership and disownership. \u003cem\u003eConsciousness and Cognition\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 82\u0026ndash;93 (2011).\u003c/li\u003e\n\u003cli\u003eBlanke, O. Multisensory brain mechanisms of bodily self-consciousness. \u003cem\u003eNat Rev Neurosci\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 556\u0026ndash;571 (2012).\u003c/li\u003e\n\u003cli\u003eBorghi, A. M. \u0026amp; Cimatti, F. Embodied cognition and beyond: Acting and sensing the body. \u003cem\u003eNeuropsychologia\u003c/em\u003e \u003cstrong\u003e48\u003c/strong\u003e, 763\u0026ndash;773 (2010).\u003c/li\u003e\n\u003cli\u003eBerti, A. \u0026amp; Rizzolatti, G. 3.3 Coding near and far space. \u003cem\u003eThe cognitive and neural bases of spatial neglect\u003c/em\u003e 119 (2002).\u003c/li\u003e\n\u003cli\u003eBerti, A. From embodying tool to embodying alien limb: sensory-motor modulation of personal and extrapersonal space. \u003cem\u003eCogn Process\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 121\u0026ndash;126 (2021).\u003c/li\u003e\n\u003cli\u003eIriki, A., Tanaka, M. \u0026amp; Iwamura, Y. Coding of modified body schema during tool use by macaque postcentral neurones. \u003cem\u003eNeuroreport\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 2325\u0026ndash;2330 (1996).\u003c/li\u003e\n\u003cli\u003eBerti, A. \u0026amp; Frassinetti, F. When Far Becomes Near: Remapping of Space by Tool Use. \u003cem\u003eJournal of Cognitive Neuroscience\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 415\u0026ndash;420 (2000).\u003c/li\u003e\n\u003cli\u003eMaravita, A. \u0026amp; Iriki, A. Tools for the body (schema). \u003cem\u003eTrends in Cognitive Sciences\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 79\u0026ndash;86 (2004).\u003c/li\u003e\n\u003cli\u003eCanzoneri, E. \u003cem\u003eet al.\u003c/em\u003e Tool-use reshapes the boundaries of body and peripersonal space representations. \u003cem\u003eExp Brain Res\u003c/em\u003e \u003cstrong\u003e228\u003c/strong\u003e, 25\u0026ndash;42 (2013).\u003c/li\u003e\n\u003cli\u003eCardinali, L. \u003cem\u003eet al.\u003c/em\u003e Tool-use induces morphological updating of the body schema. \u003cem\u003eCurrent biology\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, R478\u0026ndash;R479 (2009).\u003c/li\u003e\n\u003cli\u003eCardinali, L. \u003cem\u003eet al.\u003c/em\u003e Grab an object with a tool and change your body: tool-use-dependent changes of body representation for action. \u003cem\u003eExp Brain Res\u003c/em\u003e \u003cstrong\u003e218\u003c/strong\u003e, 259\u0026ndash;271 (2012).\u003c/li\u003e\n\u003cli\u003eRomano, D., Caffa, E., Hernandez-Arieta, A., Brugger, P. \u0026amp; Maravita, A. The robot hand illusion: Inducing proprioceptive drift through visuo-motor congruency. \u003cem\u003eNeuropsychologia\u003c/em\u003e \u003cstrong\u003e70\u003c/strong\u003e, 414\u0026ndash;420 (2015).\u003c/li\u003e\n\u003cli\u003eGarbarini, F. \u003cem\u003eet al.\u003c/em\u003e When your arm becomes mine: pathological embodiment of alien limbs using tools modulates own body representation. \u003cem\u003eNeuropsychologia\u003c/em\u003e \u003cstrong\u003e70\u003c/strong\u003e, 402\u0026ndash;413 (2015).\u003c/li\u003e\n\u003cli\u003eAymerich-Franch, L., Petit, D., Ganesh, G. \u0026amp; Kheddar, A. Embodiment of a humanoid robot is preserved during partial and delayed control. in \u003cem\u003e2015 IEEE International Workshop on Advanced Robotics and its Social Impacts (ARSO)\u003c/em\u003e 1\u0026ndash;5 (IEEE, 2015).\u003c/li\u003e\n\u003cli\u003eFarizon, D., Dominey, P. F. \u0026amp; Ventre-Dominey, J. Insights on embodiment induced by visuo-tactile stimulation during robotic telepresence. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 22718 (2021).\u003c/li\u003e\n\u003cli\u003eVentre-Dominey, J. \u003cem\u003eet al.\u003c/em\u003e Embodiment into a robot increases its acceptability. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 10083 (2019).\u003c/li\u003e\n\u003cli\u003eFarizon, D., Dominey, P. F. \u0026amp; Ventre-Dominey, J. Social Attitude Towards a Robot is Promoted by Motor-Induced Embodiment Independently of Spatial Perspective. \u003cem\u003eIEEE Robotics and Automation Letters\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 9036\u0026ndash;9042 (2022).\u003c/li\u003e\n\u003cli\u003eBotvinick, M. \u0026amp; Cohen, J. Rubber hands \u0026lsquo;feel\u0026rsquo; touch that eyes see. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e391\u003c/strong\u003e, 756\u0026ndash;756 (1998).\u003c/li\u003e\n\u003cli\u003eJeannerod, M. The representing brain: Neural correlates of motor intention and imagery. \u003cem\u003eBehavioral and Brain sciences\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 187\u0026ndash;202 (1994).\u003c/li\u003e\n\u003cli\u003eJeannerod, M. Mental imagery in the motor context. \u003cem\u003eNeuropsychologia\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 1419\u0026ndash;1432 (1995).\u003c/li\u003e\n\u003cli\u003eJeannerod, M. Neural simulation of action: a unifying mechanism for motor cognition. \u003cem\u003eNeuroimage\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, S103\u0026ndash;S109 (2001).\u003c/li\u003e\n\u003cli\u003eJeannerod, M. \u003cem\u003eMotor Cognition: What Actions Tell the Self\u003c/em\u003e. vol. 42 (OuP Oxford, 2006).\u003c/li\u003e\n\u003cli\u003edi Pellegrino, G., Fadiga, L., Fogassi, L., Gallese, V. \u0026amp; Rizzolatti, G. Understanding motor events: a neurophysiological study. \u003cem\u003eExp Brain Res\u003c/em\u003e \u003cstrong\u003e91\u003c/strong\u003e, 176\u0026ndash;180 (1992).\u003c/li\u003e\n\u003cli\u003eRizzolatti, G. \u0026amp; Craighero, L. THE MIRROR-NEURON SYSTEM. \u003cem\u003eAnnual Review of Neuroscience\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 169\u0026ndash;192 (2004).\u003c/li\u003e\n\u003cli\u003eSte-Marie, D. M. \u003cem\u003eet al.\u003c/em\u003e Observation interventions for motor skill learning and performance: an applied model for the use of observation. \u003cem\u003eInternational Review of Sport and Exercise Psychology\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 145\u0026ndash;176 (2012).\u003c/li\u003e\n\u003cli\u003eSte-Marie, D. M., Lelievre, N. \u0026amp; St. Germain, L. Revisiting the Applied Model for the Use of Observation: A Review of Articles Spanning 2011\u0026ndash;2018. \u003cem\u003eResearch Quarterly for Exercise and Sport\u003c/em\u003e \u003cstrong\u003e91\u003c/strong\u003e, 594\u0026ndash;617 (2020).\u003c/li\u003e\n\u003cli\u003eSchuch, S., Bayliss, A. P., Klein, C. \u0026amp; Tipper, S. P. Attention modulates motor system activation during action observation: evidence for inhibitory rebound. \u003cem\u003eExp Brain Res\u003c/em\u003e \u003cstrong\u003e205\u003c/strong\u003e, 235\u0026ndash;249 (2010).\u003c/li\u003e\n\u003cli\u003eWright, D. J. \u003cem\u003eet al.\u003c/em\u003e Directing visual attention during action observation modulates corticospinal excitability. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, e0190165 (2018).\u003c/li\u003e\n\u003cli\u003eDecety, J. The neurophysiological basis of motor imagery. \u003cem\u003eBehavioural brain research\u003c/em\u003e \u003cstrong\u003e77\u003c/strong\u003e, 45\u0026ndash;52 (1996).\u003c/li\u003e\n\u003cli\u003eFrak, V., Paulignan, Y. \u0026amp; Jeannerod, M. Orientation of the opposition axis in mentally simulated grasping. \u003cem\u003eExperimental Brain Research\u003c/em\u003e \u003cstrong\u003e136\u003c/strong\u003e, 120\u0026ndash;127 (2001).\u003c/li\u003e\n\u003cli\u003eJacobs, S., Danielmeier, C. \u0026amp; Frey, S. H. Human anterior intraparietal and ventral premotor cortices support representations of grasping with the hand or a novel tool. \u003cem\u003eJournal of Cognitive Neuroscience\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 2594\u0026ndash;2608 (2010).\u003c/li\u003e\n\u003cli\u003eRieger, M. \u0026amp; Massen, C. Tool characteristics in imagery of tool actions. \u003cem\u003ePsychological Research\u003c/em\u003e \u003cstrong\u003e78\u003c/strong\u003e, 10\u0026ndash;17 (2014).\u003c/li\u003e\n\u003cli\u003eLotze, M. \u0026amp; Halsband, U. Motor imagery. \u003cem\u003eJournal of Physiology-paris\u003c/em\u003e \u003cstrong\u003e99\u003c/strong\u003e, 386\u0026ndash;395 (2006).\u003c/li\u003e\n\u003cli\u003eRuffino, C., Papaxanthis, C. \u0026amp; Lebon, F. Neural plasticity during motor learning with motor imagery practice: Review and perspectives. \u003cem\u003eNeuroscience\u003c/em\u003e \u003cstrong\u003e341\u003c/strong\u003e, 61\u0026ndash;78 (2017).\u003c/li\u003e\n\u003cli\u003eMoran, A. \u0026amp; O\u0026rsquo;Shea, H. Motor imagery practice and cognitive processes. \u003cem\u003eFrontiers in Psychology\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 394 (2020).\u003c/li\u003e\n\u003cli\u003eEaves, D. L., Riach, M., Holmes, P. S. \u0026amp; Wright, D. J. Motor Imagery during Action Observation: A Brief Review of Evidence, Theory and Future Research Opportunities. \u003cem\u003eFront. Neurosci.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, (2016).\u003c/li\u003e\n\u003cli\u003eVogt, S., Di Rienzo, F., Collet, C., Collins, A. \u0026amp; Guillot, A. Multiple roles of motor imagery during action observation. \u003cem\u003eFront. Hum. Neurosci.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, (2013).\u003c/li\u003e\n\u003cli\u003eAdamo, P. \u003cem\u003eet al.\u003c/em\u003e Effects of Action Observation Plus Motor Imagery Administered by Immersive Virtual Reality on Hand Dexterity in Healthy Subjects. \u003cem\u003eBioengineering\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eConnelly, N., Welsby, E., Lange, B. \u0026amp; Hordacre, B. Virtual Reality Action Observation and Motor Imagery to Enhance Neuroplastic Capacity in the Human Motor Cortex: A Pilot Double-blind, Randomized Cross-over Trial. \u003cem\u003eNeuroscience\u003c/em\u003e \u003cstrong\u003e549\u003c/strong\u003e, 92\u0026ndash;100 (2024).\u003c/li\u003e\n\u003cli\u003eFrank, C. \u003cem\u003eet al.\u003c/em\u003e Motor imagery during action observation in virtual reality: the impact of watching myself performing at a level I have not yet achieved. \u003cem\u003eInternational Journal of Sport and Exercise Psychology\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 401\u0026ndash;427 (2023).\u003c/li\u003e\n\u003cli\u003eChoi, J. W., Kim, B. H., Huh, S. \u0026amp; Jo, S. Observing Actions Through Immersive Virtual Reality Enhances Motor Imagery Training. \u003cem\u003eIEEE Transactions on Neural Systems and Rehabilitation Engineering\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 1614\u0026ndash;1622 (2020).\u003c/li\u003e\n\u003cli\u003eLakshminarayanan, K. \u003cem\u003eet al.\u003c/em\u003e The effect of combining action observation in virtual reality with kinesthetic motor imagery on cortical activity. \u003cem\u003eFront. Neurosci.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eAlimardani, M., Nishio, S. \u0026amp; Ishiguro, H. The Importance of Visual Feedback Design in BCIs; from Embodiment to Motor Imagery Learning. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, e0161945 (2016).\u003c/li\u003e\n\u003cli\u003eOno, Y., Wada, K., Kurata, M. \u0026amp; Seki, N. Enhancement of motor-imagery ability via combined action observation and motor-imagery training with proprioceptive neurofeedback. \u003cem\u003eNeuropsychologia\u003c/em\u003e \u003cstrong\u003e114\u003c/strong\u003e, 134\u0026ndash;142 (2018).\u003c/li\u003e\n\u003cli\u003eHolper, L. \u003cem\u003eet al.\u003c/em\u003e Testing the potential of a virtual reality neurorehabilitation system during performance of observation, imagery and imitation of motor actions recorded by wireless functional near-infrared spectroscopy (fNIRS). \u003cem\u003eJ NeuroEngineering Rehabil\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 57 (2010).\u003c/li\u003e\n\u003cli\u003eBinks, J. \u003cem\u003eet al.\u003c/em\u003e Combining Action Observation and Motor Imagery (AO+MI): Mechanisms, Applications, and Guidelines for Enhancing Motor Learning and Rehabilitation. in \u003cem\u003eThe Psychology of Mental Imagery in Movement and Action\u003c/em\u003e (Routledge, 2025).\u003c/li\u003e\n\u003cli\u003eD\u0026rsquo;Angelo, M., Di Pellegrino, G., Seriani, S., Gallina, P. \u0026amp; Frassinetti, F. The sense of agency shapes body schema and peripersonal space. \u003cem\u003eScientific reports\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 13847 (2018).\u003c/li\u003e\n\u003cli\u003eGaligani, M. \u003cem\u003eet al.\u003c/em\u003e Effect of tool-use observation on metric body representation and peripersonal space. \u003cem\u003eNeuropsychologia\u003c/em\u003e \u003cstrong\u003e148\u003c/strong\u003e, 107622 (2020).\u003c/li\u003e\n\u003cli\u003eCostello, M. C. \u0026amp; Bloesch, E. K. Are older adults less embodied? A review of age effects through the lens of embodied cognition. \u003cem\u003eFrontiers in psychology\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 267 (2017).\u003c/li\u003e\n\u003cli\u003eSorrentino, G. \u003cem\u003eet al.\u003c/em\u003e How ageing shapes body and space representations: A comparison study between healthy young and older adults. \u003cem\u003eCortex\u003c/em\u003e \u003cstrong\u003e136\u003c/strong\u003e, 56\u0026ndash;76 (2021).\u003c/li\u003e\n\u003cli\u003eVerschure, P. F. M. J., P\u0026aacute;scoa dos Santos, F. \u0026amp; Sharma, V. Redefining stroke rehabilitation: Mobilizing the embodied goal-oriented brain. \u003cem\u003eCurrent Opinion in Neurobiology\u003c/em\u003e \u003cstrong\u003e83\u003c/strong\u003e, 102807 (2023).\u003c/li\u003e\n\u003cli\u003eZbinden, J., Lendaro, E. \u0026amp; Ortiz-Catalan, M. Prosthetic embodiment: systematic review on definitions, measures, and experimental paradigms. \u003cem\u003eJ NeuroEngineering Rehabil\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 37 (2022).\u003c/li\u003e\n\u003cli\u003eChye, S. \u003cem\u003eet al.\u003c/em\u003e The effects of combined action observation and motor imagery on corticospinal excitability and movement outcomes: Two meta-analyses. \u003cem\u003eNeuroscience \u0026amp; Biobehavioral Reviews\u003c/em\u003e \u003cstrong\u003e143\u003c/strong\u003e, 104911 (2022).\u003c/li\u003e\n\u003cli\u003eThe effect of action observation and motor imagery combinations on upper limb kinematics and EMG during dart‐throwing - Romano Smith - 2019 - Scandinavian Journal of Medicine \u0026amp; Science in Sports - Wiley Online Library. https://onlinelibrary.wiley.com/doi/abs/10.1111/sms.13534.\u003c/li\u003e\n\u003cli\u003eEaves, D. L., Haythornthwaite, L. \u0026amp; Vogt, S. Motor imagery during action observation modulates automatic imitation effects in rhythmical actions. \u003cem\u003eFrontiers in human neuroscience\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 28 (2014).\u003c/li\u003e\n\u003cli\u003eMeers, R., Nuttall, H. E. \u0026amp; Vogt, S. Motor imagery alone drives corticospinal excitability during concurrent action observation and motor imagery. \u003cem\u003eCortex\u003c/em\u003e \u003cstrong\u003e126\u003c/strong\u003e, 322\u0026ndash;333 (2020).\u003c/li\u003e\n\u003cli\u003eScott, M. W. An error in our thinking: The sensory conflict hypothesis of combined action observation and motor imagery. \u003cem\u003eBrain and Cognition\u003c/em\u003e \u003cstrong\u003e189\u003c/strong\u003e, 106350 (2025).\u003c/li\u003e\n\u003cli\u003eGr\u0026egrave;zes, J. \u0026amp; Decety, J. Functional anatomy of execution, mental simulation, observation, and verb generation of actions: A meta-analysis. \u003cem\u003eHuman Brain Mapping\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 1\u0026ndash;19 (2001).\u003c/li\u003e\n\u003cli\u003eH\u0026eacute;tu, S. \u003cem\u003eet al.\u003c/em\u003e The neural network of motor imagery: an ALE meta-analysis. \u003cem\u003eNeuroscience \u0026amp; Biobehavioral Reviews\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 930\u0026ndash;949 (2013).\u003c/li\u003e\n\u003cli\u003eCaspers, S., Zilles, K., Laird, A. R. \u0026amp; Eickhoff, S. B. ALE meta-analysis of action observation and imitation in the human brain. \u003cem\u003eNeuroImage\u003c/em\u003e \u003cstrong\u003e50\u003c/strong\u003e, 1148\u0026ndash;1167 (2010).\u003c/li\u003e\n\u003cli\u003eHardwick, R. M., Caspers, S., Eickhoff, S. B. \u0026amp; Swinnen, S. P. Neural correlates of action: Comparing meta-analyses of imagery, observation, and execution. \u003cem\u003eNeuroscience \u0026amp; Biobehavioral Reviews\u003c/em\u003e \u003cstrong\u003e94\u003c/strong\u003e, 31\u0026ndash;44 (2018).\u003c/li\u003e\n\u003cli\u003eDifferentiating between self and others: an ALE meta-analysis of fMRI studies of self-recognition and theory of mind | Brain Imaging and Behavior | Springer Nature Link. https://link.springer.com/article/10.1007/s11682-013-9266-8.\u003c/li\u003e\n\u003cli\u003eGallese, V. Being like me\u0026rdquo;: Self-other identity, mirror neurons and empathy. \u003cem\u003ePerspectives on imitation: From cognitive neuroscience to social science\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 101\u0026ndash;18 (2005).\u003c/li\u003e\n\u003cli\u003eSposito, A. V., Bolognini, N., Vallar, G., Posteraro, L. \u0026amp; Maravita, A. The spatial encoding of body parts in patients with neglect and neurologically unimpaired participants. \u003cem\u003eNeuropsychologia\u003c/em\u003e \u003cstrong\u003e48\u003c/strong\u003e, 334\u0026ndash;340 (2010).\u003c/li\u003e\n\u003cli\u003eLove, J. \u003cem\u003eet al.\u003c/em\u003e JASP: Graphical statistical software for common statistical designs. \u003cem\u003eJournal of Statistical Software\u003c/em\u003e \u003cstrong\u003e88\u003c/strong\u003e, 1\u0026ndash;17 (2019).\u003c/li\u003e\n\u003cli\u003eMalouin, F. \u003cem\u003eet al.\u003c/em\u003e The Kinesthetic and Visual Imagery Questionnaire (KVIQ) for Assessing Motor Imagery in Persons with Physical Disabilities: A Reliability and Construct Validity Study. \u003cem\u003eJournal of Neurologic Physical Therapy\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 20 (2007).\u003c/li\u003e\n\u003cli\u003eRozand, V., Lebon, F., Papaxanthis, C. \u0026amp; Lepers, R. Effect of mental fatigue on speed\u0026ndash;accuracy trade-off. \u003cem\u003eNeuroscience\u003c/em\u003e \u003cstrong\u003e297\u003c/strong\u003e, 219\u0026ndash;230 (2015).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Embodiment, Robotic telepresence, Motor imagery, Action observation, Body schema, Peripersonal space","lastPublishedDoi":"10.21203/rs.3.rs-9343223/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9343223/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRobotic telepresence offers a powerful framework for investigating embodiment and the plasticity of body and space representations. In this study, we examined how action observation (AO) and combined action observation with motor imagery (AOMI) modulate peripersonal space (PPS), body schema, and subjective embodiment when participants are immersed in a humanoid robot from a first-person perspective. Across two experiments using an identical telepresence setup, participants either passively observed robotic pointing movements or simultaneously observed and imagined performing the same actions. Behavioral measures assessed PPS boundaries and forearm metric representation, while questionnaires quantified ownership, self-location, and agency as markers of embodiment. Results showed that AO alone induced a contraction of PPS without altering body schema, accompanied by partial embodiment (i.e., characterized by ownership and self-location but weak agency). In contrast, AOMI produced additional changes in body schema and stronger embodiment, with enhanced agency and imagery vividness also correlating with self-location. These findings demonstrate a dissociation between spatial and bodily representations and highlight the critical role of intentional motor simulation in reshaping body schema during robotic embodiment. Overall, the study shows that embodiment in telepresence is a graded phenomenon in which motor imagery plays a key role in integrating robotic bodies into internal representations of the body self.\u003c/p\u003e","manuscriptTitle":"Embodied Motor Imagery reshapes body and space representations","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-11 05:56:28","doi":"10.21203/rs.3.rs-9343223/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"232675100032426849445042365427898617763","date":"2026-04-29T06:33:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"283860497036806308963119048485608562459","date":"2026-04-28T13:52:38+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-28T13:35:15+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-24T11:40:30+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-04-20T07:09:08+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-13T13:28:31+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-04-13T10:42:51+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f8af9e57-46f0-472e-b7a3-888066e87885","owner":[],"postedDate":"May 11th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67329747,"name":"Biological sciences/Neuroscience"},{"id":67329748,"name":"Biological sciences/Psychology"},{"id":67329749,"name":"Social science/Psychology"}],"tags":[],"updatedAt":"2026-05-11T05:56:28+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-11 05:56:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9343223","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9343223","identity":"rs-9343223","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}