RESEARCH DOSSIER
Proprioception and Body Representation
Abstract. Proprioception and body representation constitute a connected psychological and neurophysiological field concerned with how organisms estimate limb position, movement and bodily geometry, and how sensory evidence supports action and the experience of bodily ownership. This critical narrative research monograph examines the historical and physiological basis of kinesthesia, muscle-spindle afferents, position matching, implicit body metrics, visual–proprioceptive integration, predictive state estimation, rubber-hand and full-body illusions, peripersonal space, tool-induced plasticity, neural mechanisms and clinical dissociations. Original studies are evaluated alongside foundational reviews, experimental-method studies, a 2023 meta-analysis, 2024 psychometric findings and 2026 proprioceptive-integration experiments. The synthesis distinguishes sensory precision from localization bias, ownership from proprioceptive drift, and computational model fit from demonstrated neural mechanism. It closes with replicability, methodological limitations and a comparative research horizon for embodied artificial systems. The approach is an evidence-linked critical narrative synthesis, not a systematic review or original experimental investigation.
Keywords: proprioception; kinesthesia; body representation; body schema; body image; somatoperception; joint-position sense; body ownership; rubber hand illusion; proprioceptive drift; muscle spindles; multisensory integration; peripersonal space; predictive body-state estimation; Artificial Era
The Research Object: Proprioception and the Represented Body
Proprioception is the set of sensory processes through which an organism registers the configuration, movement, and mechanical state of its own body. In human experimental psychology it includes perceived joint and limb position, movement direction and speed, and the contribution of muscle-derived signals to experienced force. Body representation names the organized information that makes these signals interpretable as the position, shape, and possibilities for action of a particular body. The two concepts are functionally connected because afferent information must be interpreted against a changing configuration of body segments, postural constraints, and learned body dimensions. The physiological synthesis by Proske & Gandevia (2012) establishes the sensory foundations; the psychological analysis of Longo et al. (2010) demonstrates why bodily perception extends beyond receptor encoding.
The central object of this monograph is the transformation of proprioceptive signals into estimates of bodily geometry and lived bodily location. Three kinds of evidence are particularly informative: controlled changes to afferent input, discrepancies between what a person feels and what is physically measured, and manipulations of multisensory information that alter ownership or apparent self-location. These methods examine interacting computations while allowing distinct dependent measures. A judged hand position may shift without a parallel shift in reported ownership; a person may experience an owned limb while still using a different estimate for goal-directed movement. The experimental dissociations described by Rohde et al. (2011) and Kammers et al. (2009) make this separation central to the research design.
The guiding questions concern which receptors provide information, how the nervous system estimates joint state and body dimensions, how vision and touch recalibrate proprioception, and why explicit bodily experience can diverge from action-guiding representations. The account integrates peripheral neurophysiology, computational state estimation, psychophysics, neuropsychology, and experimentally induced body illusions. Its claims are grounded in original studies and critical reviews, with the inferential status of each approach made explicit. Functional comparisons to artificial sensorimotor systems are considered only after the human evidence is defined. The principal aim is a coherent explanation of an experimentally tractable psychological domain, rather than a new theory or a claim of exhaustive systematic-review coverage de Vignemont (2010) Blanke (2012).
Historical Formation of Kinesthesia and the Body Schema
Proprioception acquired its modern experimental precision through the study of movement sensation, muscle afferents, and the clinical consequences of lost sensory input. Earlier discussions of a bodily schema brought together observations from neurology, action, and disturbances of body awareness, but they did not establish one universally accepted architecture. Subsequent behavioral research made anatomical and functional distinctions possible by separating position judgments, perceived motion, and movement control. The classic experiments of Goodwin et al. (1972) directly altered the sensory activity available during apparently stationary or moving limbs, helping to establish muscle-derived input as a central source of kinesthetic experience.
The terminology of body schema and body image developed through intersecting clinical and psychological traditions. The former typically denotes dynamic bodily information relevant to posture and action, while the latter has been used for explicit perception, beliefs about bodily form, and affectively significant appearance. These definitions have shifted across authors, and individual tasks sometimes recruit both. The conceptual reconstruction of de Vignemont (2010) is therefore methodologically important: a dissociation only tests a proposed division when both the representation and the demand imposed by the task are clearly specified. The vocabulary must preserve differences between sensorimotor geometry, somatic perception, and evaluative body image.
Scientific history in this field is best reconstructed as a sequence of new observations and more discriminating measurements. Muscle vibration supplied an intervention into position sense; recordings from afferent fibers connected that intervention to receptor activity; visual–proprioceptive conflict made multisensory weighting measurable; and limb-ownership illusions created controlled access to the body's perceived boundaries. Work by Burke et al. (1976), van Beers et al. (1999), and Botvinick & Cohen (1998) exemplifies these transitions. The result was a research program in which the bodily subject could be studied through causal perturbations, behavioral reports, and physiological recordings rather than inferred from introspection alone. Historical terms remain useful when accompanied by an operational account of their experimental referents.
Peripheral Receptors and the Construction of Position Sense
Muscle spindles convey information related to muscle length and its changes, with primary and secondary endings contributing different sensitivities across movement and posture. Golgi tendon organs respond to muscle tension, while cutaneous and joint receptors provide additional context for bodily configuration. No single receptor class independently supplies a complete geometrical map of the limb. Muscle lengths must be interpreted with the joint structure, momentary activation, and history of contraction in view. The major review by Proske & Gandevia (2012) documents this physiological organization, and direct human recordings by Burke et al. (1976) show that vibration changes discharge in receptor populations that subsequently contribute to the interpretation of limb movement.
Several methodological complications follow from receptor physiology. Muscle thixotropy and previous contractions change spindle sensitivity, so identical physical joint angles can yield different judgments after different movement histories. The nervous system also receives sensory evidence from skin deformation and related signals that may be informative in one posture but ambiguous in another. Consequently, position-matching tasks measure the outcome of interacting sensory and decision processes, not a transparent reading of one receptor population. Reliable inference requires controlling starting positions, movement direction, recent activity, target duration, and the possibility of visual or tactile cues Proske & Gandevia (2012) van Beers et al. (1998).
Perceived muscle force introduces a further dimension. The sense of effort is influenced by central motor command, whereas experienced tension and movement arise from interacting efferent and afferent signals. Experimental reports must distinguish physical output, experienced exertion, and estimated external resistance; each can vary under fatigue or altered feedback. A synthesis that collapses them into one 'proprioceptive intensity' would lose major explanatory structure. The distinction matters in rehabilitation, exercise research, and embodied control, where a participant's reported effort may increase while objective force is unchanged or reduced. The broader sensory classification and its limitations are reviewed by Proske & Gandevia (2012), while contemporary perturbation research continues to refine the relation between muscle signals and perceived state Mortensen & Christensen (2026).
Kinesthesia, Movement Velocity, and Sensory Interventions
Kinesthesia is the experience of movement in the absence of a requirement to see the moving segment. Tendon vibration is a particularly influential manipulation because it alters the activity of muscle-spindle afferents while the body may remain stationary. The original work of Goodwin et al. (1972) established striking illusions of movement and demonstrated the critical contribution of muscle afferents under altered joint input. An illusion of moving the elbow or wrist shows that the brain interprets neural activity as evidence about limb motion, rather than directly accessing the physical fact of movement. It also creates a controlled opportunity to compare reported motion with imposed perturbation.
Human microneurographic experiments provide an important evidential bridge. In Burke et al. (1976), investigators recorded from identified endings during tendon vibration, linking the physiological perturbation to specific changes in discharge. Such data establish that the intervention can affect sensory encoding, but the resulting percept depends on spinal, cerebellar, cortical, and multisensory processing that must be studied separately. An experimental interpretation should therefore resist treating perceptual illusion magnitude as a direct numerical readout of spindle firing rate. Distinct experimental protocols differ in vibration frequency, amplitude, limb stabilization, concurrent effort, and instructions, all of which can alter the relation between peripheral input and the participant's judgment.
Recent work by Mortensen & Christensen (2026) explicitly examines the relation between perceived velocity and position during active movement under vibration. Its contribution is to test how information about ongoing velocity can enter the estimation of limb position, a relationship predicted by several dynamic estimation models. The result has significance beyond a single illusion paradigm: an accurate positional estimate may depend on integrating a movement signal over time, with uncertainty accumulating as the estimate evolves. Yet a successful model fit does not uniquely identify the neural computation or prove that one algorithm is implemented in the brain. Model comparisons require matched observables, manipulated assumptions, and an analysis of possible alternative mechanisms.
Body Geometry: From Receptors to Limb Dimensions
To locate a fingertip, the nervous system needs more than joint angles. It must also represent relationships between joints, segment lengths, hand shape, and the transformation from segment-centered signals to bodily and external coordinates. This is a geometric inference problem. A given pattern of muscle-derived input acquires spatial meaning only relative to a model of the physical structure whose state is being estimated. In the study of Longo & Haggard (2010), participants indicated landmarks on their unseen hands, allowing researchers to infer the implicit spatial organization of hand representation. Their estimates contained systematic distortions, showing that behavior can reveal a body metric different from explicit knowledge of hand appearance.
Such distortions cannot be interpreted simply as a person believing that the hand has the wrong size. Different tasks access different components of body representation. Explicit judgments of hand dimensions may be relatively accurate while landmark-based localization shows contracted fingers or expanded width; tasks also differ in external response coordinates and cognitive demands. Follow-up conceptual work by Longo & Haggard (2012) emphasizes that body representation has measurable internal structure rather than forming a unitary image. The contrast is theoretically significant because a representation specialized for rapid tactile location need not have the same coding geometry as a conscious drawing or verbal description.
A rigorous geometry experiment compares physical anatomical landmarks with judged landmarks, quantifies directional and scaling errors, and evaluates whether distortions persist under altered instructions or response methods. It should also establish which sensory evidence is available: vision of the limb, remembered shape, contact with an apparatus, or active movement. The inferential objective is to identify what an experiment operationalizes, rather than to assign every mislocalization to a single named schema. Body geometry connects proprioception with tactile spatial constancy, action planning, and the conscious experience of bodily extent Longo et al. (2010) van Beers et al. (1998).
Visual and Proprioceptive Cue Combination
An observer estimating hand location can exploit vision, proprioception, and information about the task. A central issue is how conflicting estimates are combined when they differ in precision or spatial reference frame. In van Beers et al. (1999), controlled hand-localization experiments supported a model that incorporates the precision and coordinate-dependent uncertainty of the sensory signals. Its explanatory achievement lies in deriving conditions under which combining cues should improve estimates, rather than asserting that vision always dominates. Since uncertainty varies with posture and direction, the same individual can weight the available sources differently across spatial dimensions or laboratory setups.
A simplified statistically efficient estimator combines two unbiased independent estimates by assigning weights in inverse proportion to their variances. If visual variance is small relative to proprioceptive variance, vision contributes more to the combined estimate; if visual evidence becomes uncertain, proprioception becomes more influential. Ernst & Banks (2002) demonstrated near-statistically optimal integration in a related visual–haptic judgment paradigm, providing a powerful quantitative analogy for body localization. Real sensorimotor estimates, however, may include correlated noises, prior expectations, delayed feedback, and coordinate transformations that make the elementary formula insufficient. Experiments must estimate these properties rather than assume them.
Precision and accuracy also require different analysis. A combined estimate may have low trial-to-trial variability while remaining biased by distorted body metrics or learned reference frames. A reported shift toward a seen limb is therefore not necessarily an improvement in veridical localization, and it may differ from a shift of the motor plan. Baseline measurements for vision-only and proprioception-only conditions, uncertainty estimates, and within-participant conflicts strengthen the causal interpretation of the observed integration. The complementary results of van Beers et al. (1998) and van Beers et al. (1999) show how the psychophysics of limb localization can constrain computational models without collapsing body representation into a single confidence-weighted coordinate.
Predictive State Estimation and Internal Models
The body changes during the interval required for sensory information to travel and be processed. A motor command therefore operates on a body whose current configuration cannot always be recovered from incoming signals alone. Predictive state estimation combines information about recent state, motor commands, and sensory feedback to estimate an evolving position and velocity. In Wolpert et al. (1995), movements under externally imposed forces provided evidence compatible with the use of an internal model in sensorimotor integration. The experimental logic did not require the model to be a conscious description of mechanics; it treated prediction as a computation necessary for accurate behavior under delayed and uncertain feedback.
State estimation and motor control are interdependent but analytically distinct. A forward model predicts the sensory or mechanical consequences of a command; an inverse model may map desired effects toward command requirements. Prediction errors can update expectations, while motor performance errors may also recruit explicit correction and strategy. The significance of proprioception here is its contribution to measured and predicted body state, especially when visual feedback is absent. Sainburg et al. (1999) provides evidence that anticipatory, corrective, and postural mechanisms contribute at different phases of reaching, illustrating why a single behavioral deviation cannot identify a unique internal computation.
Formal approaches such as Bayesian filtering and optimal feedback control specify relationships among uncertain state estimates, dynamics, and task costs. Their use is valuable insofar as they generate contrasting predictions about intervention timing, response variability, and compensation for unexpected loads. Yet different models can predict similar movement trajectories, and their latent states may be difficult to identify from behavior alone. Independent constraints from proprioceptive reports, afferent perturbation, neural activity, and generalization tests increase discriminability. The broad methodological critique of Krakauer et al. (2019) also cautions against treating short-term recalibration as evidence for every type of learned motor skill.
Body Schema, Body Image, and Somatoperception
Body schema commonly refers to an action-sensitive and continuously updated organization of bodily configuration, whereas body image commonly refers to explicit bodily knowledge and perceptual experience, sometimes including affective evaluation. The distinction is useful only when the investigator specifies the observable task and representation under discussion. de Vignemont (2010) systematically examines the merits and difficulties of dividing body representations into these categories. Some performances are automatically guided without requiring conscious report, while other judgments involve stored knowledge of body structure. Neither label names a single anatomically isolated module with a universally accepted border.
An influential finer-grained vocabulary distinguishes somatoperception, the organization of perceptual information about the body, from somatorepresentation, higher-order and longer-lived knowledge about it. Longo et al. (2010) reviews how primary sensory signals are remapped into coordinates suitable for perceiving objects contacting the skin, identifying bodily states, and locating stimulation. This approach explains why changing the response task can change what appears to be a bodily representation. A participant asked to point to the knuckle, imagine a hand outline, or execute a reach engages overlapping sensory information but different transformations and goal demands.
The practical consequence is a multidimensional research architecture. Limb posture, structural metrics, ownership, self-location, motor possibilities, and evaluative body beliefs may covary, but each needs its own operational test. Researchers should state whether the hypothesized representation is perceptual, sensorimotor, semantic, or affective, and specify when it is updated. Comparisons across tasks are strongest when sensory conditions and response mappings are matched and differences can be assigned to defined requirements. Longo & Haggard (2012) suggests that the unity of bodily experience can coexist with dissociable underlying representations, a conclusion that organizes the monograph's subsequent analysis of illusion, action, and clinical disorders.
Implicit Body Metrics and Conscious Knowledge
Implicit hand mapping demonstrates how a body representation can remain systematically nonveridical while everyday action is largely functional. In Longo & Haggard (2010), participants judged the locations of anatomical landmarks of an unseen hand; reconstructed maps showed substantial and structured distortion. Such data are compelling because they measure bodily geometry without requiring the participant to choose among verbal descriptions. They also reveal a methodological problem: pointing to remembered or felt landmarks involves both the represented source position and the response coordinate system, so measured error contains contributions that may be separable only with additional experimental controls.
Explicit body-form knowledge can differ from the estimates elicited by a perceptual localization task. Longo & Haggard (2012) argues that the structure of conscious body experience is open to psychometric and psychophysical decomposition; the contrast between implicit and explicit measures is evidence for task-dependent coding. It does not entail that one task accesses a fixed 'true' body model and another a false one. In a stable perceptual system, a representation may prioritize discrimination, posture updating, or spatial constancy over an anatomically perfect metric. The empirical question is whether a proposed task-specific function predicts the magnitude, pattern, and generalization of an observed distortion.
These experiments also distinguish the physical body from its represented coordinate geometry. A remembered hand drawing can be approximately accurate because it accesses learned visual and semantic information; a tactile localization may be biased because its reference frame is linked to nonuniform somatosensory coding. Reconciling these observations requires measuring the circumstances in which the representations interact. Experimental manipulations of posture, contact location, and vision can reveal whether the same distortion affects action. The framework developed by Longo et al. (2010) connects these findings to more general somatosensory remapping rather than reducing them to artistic or verbal mistakes.
Coordinate Transformations and Multisensory Body Space
Proprioceptive encoding is distributed over receptors embedded in muscles and other tissues, while useful behavioral goals are often specified in external visual coordinates. To touch a seen point, an observer must relate retinal and eye-centered signals to shoulder, elbow, and hand coordinates and update these transformations as posture changes. The review by Macaluso & Maravita (2010) describes visual–tactile interactions that support representations of space near the body. These operations connect somatic information to the external world and can produce systematic conflicts when a seen limb is displaced relative to its actual location.
Neural recordings in primates place limits on purely introspective accounts. Graziano et al. (2000) reported coding of arm location influenced by vision, demonstrating a route by which the observed position of a limb contributes to multisensory spatial representation. This evidence supports functional convergence, although a particular cellular population cannot be equated directly with the participant's conscious ownership experience. Likewise, peripersonal spatial coding organizes interactions with nearby objects and potential contacts but does not by itself prove that every represented object is experienced as bodily. The distinction becomes especially important when researchers discuss tools, virtual avatars, and prosthetic devices.
Coordinate transformations may be modulated by task demands, delays, and reliability. An estimate useful for planning a grasp before movement may differ from the online signal useful for correcting its trajectory. Error should therefore be assessed in both proprioceptive and external spatial coordinates when possible, with explicit recording of body configuration. van Beers et al. (1999) shows that localization uncertainty is geometrically structured, and Macaluso & Maravita (2010) emphasizes that integration depends on spatial relationships. A complete theory should explain both why coordinated estimates emerge under ordinary conditions and why their breakdown produces measurable drift, altered referral of touch, or inappropriate movements.
Joint-Position Matching and Psychophysical Measurement
Position matching is a family of tasks in which the participant reproduces a target joint angle or identifies the spatial position of an unseen limb. Ipsilateral matching can minimize interlimb transformations but may introduce memory and movement-history effects; contralateral matching requires a coordinate transformation between limbs. Measures include constant error, absolute error, directional bias, variability, and dependence on limb orientation. van Beers et al. (1998) demonstrated that the precision of proprioceptive localization can be estimated across spatial directions, making uncertainty an experimental quantity rather than an informal description of felt accuracy. A high-quality design specifies whether the target is passive or active, whether movement is allowed during judgment, and how reference information is removed.
A matching error combines peripheral sensory noise with memory, attention, response selection, and mechanics of the matching movement. Repeated positions can be learned during the experiment; mechanical stops and residual tactile signals may substitute for proprioception. Therefore a joint-position test is not automatically a pure measure of spindle function. Investigators should counterbalance targets, control starting positions, monitor fatigue, and report trial-level variability and test–retest reliability. When comparing populations, a larger error may arise from a cognitive or motor difference rather than a specific peripheral sensory deficit. Combining matching with passive discrimination or independent movement detection improves the specificity of inference Proske & Gandevia (2012) van Beers et al. (1999).
Proprioceptive measurement has no single universally sufficient gold standard because tasks interrogate different components of position, motion, and force perception. A protocol should be chosen for the mechanism being tested and accompanied by construct validity evidence. The distinction between a localization endpoint and the felt state of a joint is especially relevant when studies compare experimental effects across laboratories. Precision may change with posture even if mean bias does not, and a participant may improve in one response modality while remaining unchanged in another. Scientific reports should retain the axes, units, coordinate system, and available feedback that define the measurement van Beers et al. (1998) de Vignemont (2010).
Tendon Vibration, Deafferentation, and Causal Inference
Tendon vibration perturbs a physiologically defined part of the information that normally supports kinesthesia. Classic experiments by Goodwin et al. (1972) induced movement illusions through vibration and conditions affecting joint afferents, while Burke et al. (1976) recorded discharge changes in human muscle spindle endings. Together these approaches link an experimentally controlled peripheral intervention to a perceptual consequence. The link is stronger than a mere association between subjective movement and brain activity, because the input can be manipulated. Yet the intervention may recruit multiple endings and central adjustments, so its effect does not isolate a single cortical representation or determine exactly how participants combine sensory evidence.
Loss of large-fiber proprioceptive input provides a different form of evidence. People with severe sensory deafferentation can retain motor commands but have difficulty controlling and knowing limb position when vision is unavailable. The clinical pattern illustrates how proprioception stabilizes behavior across changing postures, while compensatory use of visual information demonstrates the flexibility of sensorimotor strategies. Generalization from a particular neurological patient must remain conditional on the extent of the lesion, preserved tactile and vestibular input, disease course, and training history. Reviews such as Proske & Gandevia (2012) and de Vignemont (2010) provide the anatomical and conceptual framework needed to interpret these cases.
Experimental and clinical evidence are complementary. Acute perturbation identifies the immediate dependence of perception on an input; chronic loss reveals adaptation, compensation, and the consequences of missing input over time. Long-term patients cannot function as simple replicas of a short-lived laboratory perturbation, because neural reorganization and learned strategies alter their performance. Conversely, normal participants under vibration do not necessarily reproduce the profound effects of peripheral neuropathy. Strong mechanistic claims should specify whether they concern sensory encoding, multisensory estimation, motor correction, or consciously reported kinesthesia, and show which intervention distinguishes the possibilities Mortensen & Christensen (2026) Sainburg et al. (1999).
Rubber-Hand Illusion and the Attribution of Ownership
The rubber-hand illusion provides a controlled means of altering the apparent relationship between a body and an object. In the landmark study Botvinick & Cohen (1998), correlated tactile stimulation of a visible artificial hand and an unseen real hand could make a participant experience touch as originating from the seen hand and judge the real hand to lie closer to it. The manipulation brings vision, touch, and proprioception into measurable conflict. It transforms body ownership from an exclusively descriptive phenomenon into a research variable, although the operational measurements are related indicators rather than direct readings of an otherwise inaccessible single state.
Later work investigated the constraints on this transformation. Tsakiris & Haggard (2005) examined anatomical plausibility, posture, and temporal matching, showing that correlated stimulation operates within expectations about what may belong to the body. A plastic object and a hand-shaped object need not produce equivalent experiences; a realistic hand in an implausible location can also fail to elicit the same response. Such conditions motivate models in which incoming sensory signals are evaluated against a prior or stored bodily model. They also complicate a purely correlation-based account because the nervous system must decide which sensory events belong together in the first place.
Neuroimaging provides convergent evidence without converting a subjective report into a uniquely localized neural event. Ehrsson et al. (2004) found premotor activity associated with ownership experience during a manipulated-hand paradigm. Tsakiris (2010) later proposed a distributed neurocognitive account involving anatomical constraints, multisensory recalibration, and bodily attribution. These studies establish plausible neural involvement and model structure, while the relation between a regional signal, causal necessity, and phenomenal experience requires independent intervention and replication. The scientific outcome is a family of experimentally supported mechanisms and boundary conditions rather than one invariant brain site for bodily ownership.
Proprioceptive Drift, Embodiment, and Psychometric Dissociation
Proprioceptive drift describes a change in the judged position of a hidden limb, often toward a visible fake limb after experimental induction. It is a behavioral outcome with an interpretable spatial unit, but it does not directly measure the strength of an ownership experience. Rohde et al. (2011) reported a dissociation between subjective ownership and drift in rubber-hand conditions. Kammers et al. (2009) also examined the relation between illusion-related perception and action, adding evidence that the measures can depend on different task demands. These findings require experiments to specify which component of embodiment is their primary endpoint, rather than substitute one convenient index for another.
A quantitative synthesis in Tosi et al. (2023) found that drift and subjective embodiment are positively correlated on average, with a reported association around 0.35. The magnitude indicates shared variance while leaving substantial differences across trials, studies, and measurement methods. This result both qualifies an absolute dissociation and rejects treating the two outcomes as interchangeable. The meta-analysis is more informative than selecting a single positive or null study, although the conclusion still depends on the comparability of designs and operationalizations included. Studies that manipulate timing, distance, expectancy, or response method may generate different correlations because they influence one measured dimension more strongly.
Measurement quality remains a contemporary concern. Tosi et al. (2024) used exploratory graph analysis to examine responses about ownership, referral of touch, and control statements, finding related but potentially separable aspects and raising questions about traditional questionnaire composites. Furthermore, Lanfranco et al. (2023) applied signal detection methods to characterize sensitivity and bias during ownership judgments, demonstrating that psychophysical discrimination can complement ratings. Together these studies support a multimethod strategy: separately record subjective ownership, tactile referral, limb localization, response criterion, and movement behavior. An experiment should report which indicators agree and explain their discrepancies rather than conceal them in a combined 'illusion score'.
Full-Body Illusions, Perspective, and Self-Location
Ownership of a hand is one component of bodily self-experience, whereas the perceived location of the person and the spatial direction of first-person perspective are additional components. In Lenggenhager et al. (2007), virtual-reality arrangements created conflicting visual and somatosensory evidence about the location of a body and yielded changes in experienced self-location. Ehrsson (2007) reported an experimental illusion associated with a displaced first-person perspective. These interventions show that body-related multisensory processing can influence where a participant experiences being located, rather than only which limb is experienced as belonging to the body.
Whole-body manipulations rely on features that may matter less in hand illusions, including visual perspective, spatial scale, posture, and synchrony across extended body surfaces. Petkova & Ehrsson (2008) investigated conditions under which participants experienced a full-body substitution, making an important experimental connection between multisensory contingencies and perceived bodily identity. Blanke (2012) reviews the resulting three-component framework of self-identification, self-location, and first-person perspective. A strong interpretation keeps these dependent variables distinct because a reported feeling of owning a virtual body need not entail a comparable displacement of self-location or an equally altered visual perspective.
Experimental research on self-location must also control demand characteristics, visual geometry, motor constraints, and what participants mean by 'being' somewhere. Self-report is informative about subjective phenomenology, while behavioral localization tasks provide a different kind of evidence. Neural correlations add a further level of explanation but do not eliminate the interpretive challenge. In clinical out-of-body experiences, vestibular and neurological factors may differ substantially from those in controlled virtual conditions; analogy is a hypothesis, not an equivalence. The convergence of Blanke (2012) and Lenggenhager et al. (2007) provides a framework for relating proprioception to the larger organization of bodily self-consciousness.
Parietal, Premotor, Somatosensory, and Integrative Circuits
The neural organization of bodily perception involves interacting regions rather than a single representational center. Primary somatosensory pathways convey information about bodily stimulation and peripheral state, while posterior parietal and premotor regions participate in coordinate transformation and multisensory integration. Graziano et al. (2000) provided physiological evidence that arm-location coding can be influenced by sight, and Ehrsson et al. (2004) linked premotor activity to induced ownership experience. Reviews by Longo et al. (2010) and Tsakiris (2010) integrate complementary evidence concerning somatic perceptual constancy, bodily attribution, and the transition from localized sensory evidence to more complex bodily representations.
A regional activation is evidence of task-related neural participation, conditional on the experiment's design and the contrast selected. It is insufficient on its own to establish that a region uniquely causes conscious ownership or implements a particular numerical algorithm. The same area may contribute to spatial attention, tactile prediction, motor planning, and perceptual reports under different conditions. Kammers et al. (2009) illustrates how disturbing parietal processing with transcranial magnetic stimulation can supply more causal information, although the intervention's spatial selectivity and downstream effects still require interpretation. Concordance across lesions, stimulation, physiological recording, and behavior strengthens mechanistic inference.
Scientific modeling must connect neural activity to an explicitly defined representational variable. A study claiming coding of limb position should test whether activity follows the physical limb, the seen limb, the planned action, or an internal estimate when these are experimentally separated. A study of body ownership should distinguish neural effects of sensory synchrony from those tied to participant reports. The synthesis of Blanke (2012) underscores the contribution of temporoparietal, parietal, and premotor networks to experimentally modulated bodily self-experience. Neural observations gain explanatory force when their temporal and spatial properties predict behavior across independent perturbations.
Action Control and Dissociations Between Perception and Movement
Action-related body representations are evaluated by their contribution to successful movement, particularly under delay, uncertainty, and perturbation. In Wolpert et al. (1995), participants' estimates of hand location under imposed forces were informative about predictive sensorimotor computation. Sainburg et al. (1999) differentiated anticipatory, corrective, and postural components in reaching. These studies support a functional distinction between the information that guides an unfolding movement and a participant's explicit judgment after the event. An action can remain appropriately corrected while a verbal report of the executed trajectory is inaccurate, because the two tasks impose different access and timing requirements.
The behavioral findings of Fourneret & Jeannerod (1998) offer a clear example. Participants adjusted movements while viewing manipulated cursor feedback and underestimated the deviation in their actual hand trajectories. The experimental design indicates that online control and conscious awareness of movement can diverge under specified conditions. This pattern does not mean that action is wholly unconscious or unrelated to proprioception; it identifies a dissociation between measured action correction and a later report. Kammers et al. (2009) supplies another task-based comparison by examining how body-ownership illusions affect perceptual and movement measures differently.
Motor adaptation presents a related problem because changed behavior may arise from implicit recalibration, deliberate strategy, reinforcement, or multiple processes operating together. Krakauer et al. (2019) reviews the limitations of equating elementary sensorimotor adaptation with the acquisition of complex skill. To determine whether proprioceptive remapping has occurred, investigators can test its persistence without vision, its generalization across limbs or postures, and whether perceptual localization shifts independently of motor aiming. The logical connection between body representation and action is robust, but the mapping from a particular observed aftereffect to a specific computational mechanism remains a problem for experimental design.
Tool Use, Peripersonal Space, and Body Plasticity
Using a tool changes the spatial relation between the body and reachable objects. Studies of peripersonal space investigate how visual and tactile information are organized near a limb or tool; studies of body schema examine whether the action-guiding representation of bodily geometry itself changes. Maravita & Iriki (2004) reviews physiological and behavioral evidence for tool-related plasticity in such representations. The conceptual distinction matters because an expanded space of effective action does not necessarily imply that the tool is experienced as flesh or that conscious ownership has changed. Reachability, multisensory receptive fields, and perceived limb length should be analyzed as separable endpoints.
In Cardinali et al. (2009), using a mechanical grabber influenced subsequent free-hand movement and aspects of perceived arm geometry. The finding supports a body-related aftereffect extending beyond active tool use, making it more specific than a report that a person felt skilled with an implement. Yet alternative explanations should be considered, including movement recalibration and changes in strategy that might influence post-tool performance. The study's significance lies in providing testable predictions about how action transforms bodily coding, not in proving that all tools become permanent members of the represented body.
Peripersonal representation also changes with movement goals, threat, attention, and the relative reliability of visual and tactile evidence. Macaluso & Maravita (2010) reviews the spatial constraints on visual–tactile integration, while Makin et al. (2008) situates artificial-hand illusions in the wider context of near-body space. These mechanisms offer a framework for evaluating prostheses, remote manipulators, and avatars without presuming ownership from successful use. Experimental designs should compare task performance, explicit ownership judgments, spatial localization, and retention after the tool is removed. The central question is which bodily or action-related variable changes, over what timescale, and under which conditions.
Clinical Dissociations, Phantom Limbs, and Rehabilitation
Neurological and peripheral sensory disorders provide important tests of how the body is represented when input is altered or absent. Deafferentation can disrupt position knowledge and motor coordination, while central lesions can change attention to, localization of, or identification with a body part. Phantom-limb phenomena show that an experienced limb may persist when the physical limb is absent, raising questions about stored body models and plastic cortical representations. These observations must be distinguished from one another clinically; the same subjective description may have different sensory and neural causes. Proske & Gandevia (2012) clarifies peripheral pathways, and de Vignemont (2010) examines how clinical dissociations constrain distinctions among body representations.
Post-amputation reorganization is empirically complex. The review by Makin & Flor (2020) critically examines relationships among changed cortical representation, phantom experience, and phantom-limb pain, warning against interpreting a simple remapping narrative as a universal causal explanation. Pain intensity, movement imagery, remaining afferent input, and prosthetic experience can vary substantially across patients. Cross-sectional associations cannot alone determine whether reorganization caused pain, followed it, or reflects adaptation. Longitudinal evidence and causal perturbations are especially valuable, and patient-specific conclusions require clinically appropriate investigation rather than extrapolation from a brief illusion experiment.
Rehabilitation research should measure both task-specific improvement and meaningful functioning. Better joint-position reproduction, for example, need not automatically predict improved walking, reaching, or daily independence; altered illusion ratings cannot be treated as a validated therapeutic endpoint. Research designs benefit from standardized procedures, blinded outcome assessment where feasible, patient-relevant performance measures, and documentation of compensatory strategies. This monograph offers a scientific interpretation of evidence, not patient-level diagnosis or treatment directions. The experimental literature reviewed in Krakauer et al. (2019) and Makin & Flor (2020) shows why plasticity, motor learning, clinical outcomes, and the experience of a body should be evaluated with different but coordinated tools.
Development, Experience, and Contextual Variation
Body representations emerge through recurring correspondences among movement, vision, touch, and consequences in the environment. Development therefore concerns both maturing sensory pathways and the progressive organization of learned cross-modal relationships. A child's performance on a localization or ownership task reflects attention, instruction comprehension, response ability, and the available sensorimotor history as well as bodily representation. The influential models in Tsakiris (2010) and Longo & Haggard (2012) suggest that stored anatomical expectations interact with current sensory evidence, providing hypotheses about how experience might shape the conditions under which a seen body part is accepted as one's own.
Adult perception likewise varies with posture, expertise, fatigue, sensory history, and the immediate configuration of the task. The reliability-weighting framework in van Beers et al. (1999) predicts that changed uncertainty can alter cue use without requiring a stable personality difference or lasting rearrangement of body schema. A large difference in illusion susceptibility across groups may also reflect expectation, available visual feedback, or response scaling. To establish developmental or cultural generality, researchers need comparable instruments, sufficiently diverse samples, and measurement invariance. Findings from predominantly young laboratory volunteers should not silently be projected onto children, older adults, or clinical populations.
Learning across the lifespan can recalibrate an estimate while preserving enduring structural constraints. A temporary mismatch between seen and felt hand location may change trial-by-trial judgments; repeated use of an extended tool may change action geometry; chronic sensory deprivation may produce longer-term compensatory organization. These are different temporal processes, even if the word 'plasticity' applies to each. An adequate developmental theory should specify acquisition, retention, transfer, and recovery after perturbation. The original studies of Cardinali et al. (2009) and the motor-learning synthesis of Krakauer et al. (2019) offer contrasting paradigms for analyzing such time-dependent changes.
Contemporary Experimental Findings and Their Limits
Recent work has shifted the field toward formal measurement of the relations among perceived position, inferred movement, ownership, and response criteria. Mortensen & Christensen (2026) tested whether velocity-related proprioceptive information influences perceived position during active movement. The design builds on earlier vibration interventions while connecting them to computational predictions about temporal integration. Its strength is the explicit relationship between experimentally altered input and measured estimates; its scope is the studied movement configuration and participants. Generalization to every limb, age, neurological condition, or natural movement requires further investigation. The paper demonstrates how old physiological questions can yield new model-discriminating experiments.
Another 2026 experiment examined how delayed-onset muscle soreness affects body-related processing. In Scotto et al. (2026), a controlled comparison included an exercise-induced soreness group and a control group, each with 47 young sports science students, and tested mental rotation performance following the intervention. The reported effects were interpreted as altered efficiency of proprioceptive integration and motor imagery under a temporarily changed bodily condition. The design is informative because it links short-term peripheral or musculoskeletal disturbance with cognitive performance. Yet soreness, altered force production, fatigue, and sensory change may covary, so a single group difference does not by itself identify spindle damage as the unique causal mediator.
Research on ownership has also moved beyond traditional rating comparisons. The signal-detection analysis of Lanfranco et al. (2023) separated response bias from sensitivity when judging embodied-hand conditions, including effects of temporal asynchrony. The meta-analysis of Tosi et al. (2023) quantified the positive but moderate relation between drift and subjective embodiment, while the psychometric work of Tosi et al. (2024) questioned simple composite ownership scales. These investigations converge on a methodological conclusion: a well-defined perceptual construct needs a measure that distinguishes sensitivity, criterion, and related experiences. More elaborate statistics improve interpretation only when the assumptions and task mappings are scientifically appropriate.
Evidence Quality, Replication, and Experimental Design
The strongest claims about proprioception are supported by convergent methods that intervene at different points in the causal chain. Microneurography and vibration connect peripheral encoding to perceptual effects; controlled sensory conflicts test cue combination; action perturbations reveal predictive control; and neural or clinical studies constrain candidate substrates. Each method has limitations. Electrophysiology may reveal afferent activity without establishing its conscious meaning, a questionnaire may measure expectation as well as ownership, and observed neural activation may reflect competing task components. The evidence review should identify the causal variable and dependent measure of each study rather than summarize them all as equivalent demonstrations of embodiment Burke et al. (1976) Botvinick & Cohen (1998) Ehrsson et al. (2004).
Replication requires attention to body posture, hand separation, stimulation timing, apparatus geometry, previous movement, visual feedback, participant instructions, and sample selection. A nominally identical rubber-hand procedure may differ sufficiently in actual spatial and temporal conditions to generate divergent results. Evidence of moderation should be reported in relation to those conditions. Rohde et al. (2011) provides a key experimental dissociation, while Tosi et al. (2023) integrates multiple datasets and reports a mean correlation that makes blanket generalizations about independence untenable. Both belong in the same scientific account. Contradictory outcomes can refine a construct when their measures and designs are considered in common terms.
Statistical evaluation should distinguish within-participant change from between-person variability, account for correlated trials and repeated comparisons, and report uncertainty intervals around effects when available. Pre-registration, independent datasets, sensitivity analyses, and openly specified exclusions improve interpretability. A positive ownership rating alone is insufficient to establish a shift of anatomical representation; a changed position estimate alone does not prove that the participant identified with a fake hand. The psychometric findings of Tosi et al. (2024) and the sensitivity–bias separation in Lanfranco et al. (2023) indicate concrete routes for improved measurement. The present monograph is a critical narrative synthesis and does not claim to have performed a PRISMA systematic review or statistical meta-analysis.
Competing Explanations and Open Theoretical Problems
Several explanatory traditions describe body representation at different levels. Reliability-weighted cue integration asks how uncertain signals are combined Ernst & Banks (2002) van Beers et al. (1999). Predictive state-estimation models ask how sensory feedback and motor predictions track a changing body Wolpert et al. (1995). Anatomically constrained ownership models ask which multisensory configurations are judged to belong to one's body Tsakiris (2010). Phenomenological and neurocognitive approaches distinguish ownership, self-location, and perspective Blanke (2012). Their variables partially overlap, but they generate different predictions about what should change under an intervention. An integrated account should map those predictions onto measurable behaviors rather than select one vocabulary to describe all findings.
Outstanding questions include the stability of implicit anatomical metrics, the relation between conscious ownership and action, the dynamics of estimation under prolonged sensory uncertainty, and the degree to which peripheral input can be substituted by vision or other feedback. Another problem is model identifiability: two theories may explain the same direction of proprioceptive drift while differing about the hidden computations that produced it. Discriminating experiments should manipulate the conditions under which predictions diverge, then measure multiple endpoints and test out-of-sample generalization. The distinct conclusions of Rohde et al. (2011), Tosi et al. (2023), and Lanfranco et al. (2023) exemplify the value of progressively more informative tasks.
The relation between structural body knowledge and sensorimotor representations also remains incompletely understood. The implicit distortion reported by Longo & Haggard (2010) suggests that an anatomically useful representation may not have a fully veridical geometry, while tool aftereffects studied by Cardinali et al. (2009) show that represented bodily extent can be temporarily recalibrated. A theory must explain when such coding is stable, when it changes, and what functional advantage a distortion could confer. Clinical evidence raises the parallel question of how bodily experience persists or transforms after sensory loss or amputation Makin & Flor (2020). These problems invite coordinated experimental, computational, and longitudinal research with clear constructs and transparent limits.
The Artificial Era and Comparative Embodied Systems
Artificial sensorimotor systems provide a comparative setting for studying how estimates of body geometry emerge from sensors, learned dynamics, and adaptive control. A robot with joint encoders and cameras may require transformations among actuator coordinates, visual space, and an internal kinematic model to manipulate objects. Such computations are functionally comparable to certain aspects of proprioceptive estimation, but their comparison must specify shared operations and observable performances. The human findings in van Beers et al. (1999) and Wolpert et al. (1995) offer rigorous behavioral targets for artificial systems that combine uncertain signals and predictions, including responses to imposed delays or conflicting positions.
Functional comparison can become experimentally productive when models are evaluated with matched tasks. An artificial controller can be tested for joint-position uncertainty, cue reweighting, performance under sensory perturbation, or persistence of calibration after a tool change. These tests can illuminate computational constraints independently of attributing human-like experience. Meanwhile, the human psychology of wearable devices and embodied interfaces remains a separate empirical field: ownership questionnaires, self-location judgments, and motor performance may respond differently to a virtual or prosthetic body. The results of Petkova & Ehrsson (2008), Makin et al. (2008), and Cardinali et al. (2009) guide selection of measures while leaving the interpretation of subjective experience tied to actual human reports.
Within the Artificial Era, proprioception and body representation can therefore anchor a comparative program connecting embodied cognition, cybernetics, robotics, and psychology of human–artificial relations. The conceptual contribution is a disciplined research framework: define the system, sensory channels, coordinate geometry, adaptive processes, and behavioral criteria before comparing forms of organization. A computational analogy does not establish the presence of conscious ownership in an artificial system, while absence of a familiar biological receptor does not by itself settle whether comparable state-estimation functions are implemented. The primary scientific obligation remains experimental specificity, supported by published human research and transparent artificial-system benchmarks.
Conclusion
Proprioception supplies evolving information about bodily position, movement, and mechanical state; body representation organizes this information into estimates of geometry, action possibilities, ownership, and bodily location. The major empirical traditions converge on the view that bodily experience depends on the interaction of sensory encoding, learned anatomical structure, prediction, and multisensory inference. Physiological studies of muscle afferents Goodwin et al. (1972) Burke et al. (1976), body-localization experiments van Beers et al. (1999), and implicit hand-geometry findings Longo & Haggard (2010) collectively establish why neither raw sensation nor an isolated conscious image can account for the full phenomenon. These traditions describe different stages and uses of bodily information that require coordinated investigation.
Body-ownership research reveals especially clearly the multidimensional character of embodied experience. The original rubber-hand manipulation Botvinick & Cohen (1998) was extended by studies of anatomical expectations Tsakiris & Haggard (2005), neural involvement Ehrsson et al. (2004), whole-body self-location Lenggenhager et al. (2007), and methodological dissociation between ownership and drift Rohde et al. (2011). Contemporary meta-analytic and psychometric research Tosi et al. (2023) Tosi et al. (2024) refines the picture: related indicators share information while retaining distinct measurement meanings. The resulting scientific synthesis is stronger when each indicator is linked to a defined mechanism or construct, and when disagreements among experiments are treated as evidence about boundary conditions.
The next generation of research will benefit from combined afferent perturbation, formal state estimation, carefully validated subjective measures, causal neural interventions, and longitudinal observation in diverse populations. Human–artificial comparative experiments can add tests of computational generality, provided that function, representation, and experienced ownership remain operationally distinct. This dossier establishes proprioception and body representation as a unified field of interacting research questions with multiple measurable levels: peripheral signals, computational body-state estimates, action-guiding geometry, bodily self-experience, and adaptation. Its central conclusion is that scientifically productive integration depends on explicit definitions and discriminating tests that connect these levels without erasing their differences.
References
- Blanke, O. (2012). Multisensory brain mechanisms of bodily self-consciousness. Nature Reviews Neuroscience, 13(8), 556–571. https://doi.org/10.1038/nrn3292
- Botvinick, M., & Cohen, J. (1998). Rubber hands ‘feel’ touch that eyes see. Nature, 391(6669), 756. https://doi.org/10.1038/35784
- Burke, D., Hagbarth, K.-E., Löfstedt, L., & Wallin, B. G. (1976). The responses of human muscle spindle endings to vibration of non-contracting muscles. The Journal of Physiology, 261(3), 673–693. https://doi.org/10.1113/jphysiol.1976.sp011580
- Cardinali, L., Frassinetti, F., Brozzoli, C., Urquizar, C., Roy, A. C., & Farnè, A. (2009). Tool-use induces morphological updating of the body schema. Current Biology, 19(12), R478–R479. https://doi.org/10.1016/j.cub.2009.05.009
- de Vignemont, F. (2010). Body schema and body image—Pros and cons. Neuropsychologia, 48(3), 669–680. https://doi.org/10.1016/j.neuropsychologia.2009.09.022
- Ehrsson, H. H., Spence, C., & Passingham, R. E. (2004). That's my hand! Activity in premotor cortex reflects feeling of ownership of a limb. Science, 305(5685), 875–877. https://doi.org/10.1126/science.1097011
- Ehrsson, H. H. (2007). The experimental induction of out-of-body experiences. Science, 317(5841), 1048. https://doi.org/10.1126/science.1142175
- Ernst, M. O., & Banks, M. S. (2002). Humans integrate visual and haptic information in a statistically optimal fashion. Nature, 415(6870), 429–433. https://doi.org/10.1038/415429a
- Fourneret, P., & Jeannerod, M. (1998). Limited conscious monitoring of motor performance in normal subjects. Neuropsychologia, 36(11), 1133–1140. https://doi.org/10.1016/S0028-3932(98)00006-2
- Goodwin, G. M., McCloskey, D. I., & Matthews, P. B. C. (1972). The contribution of muscle afferents to kinaesthesia shown by vibration induced illusions of movement and by the effects of paralysing joint afferents. Brain, 95(4), 705–748. https://doi.org/10.1093/brain/95.4.705
- Graziano, M. S. A., Cooke, D. F., & Taylor, C. S. R. (2000). Coding the location of the arm by sight. Science, 290(5497), 1782–1786. https://doi.org/10.1126/science.290.5497.1782
- Kammers, M. P. M., de Vignemont, F., Verhagen, L., & Dijkerman, H. C. (2009). The rubber hand illusion in action. Neuropsychologia, 47(1), 204–211. https://doi.org/10.1016/j.neuropsychologia.2008.07.028
- Kammers, M. P. M., Verhagen, L., Dijkerman, H. C., Hogendoorn, H., de Vignemont, F., & Schutter, D. J. L. G. (2009). Is this hand for real? Attenuation of the rubber hand illusion by transcranial magnetic stimulation over the inferior parietal lobule. Journal of Cognitive Neuroscience, 21(7), 1311–1320. https://doi.org/10.1162/jocn.2009.21095
- Krakauer, J. W., Hadjiosif, A. M., Xu, J., Wong, A. L., & Haith, A. M. (2019). Motor learning. Comprehensive Physiology, 9(2), 613–663. https://doi.org/10.1002/cphy.c170043
- Lanfranco, R. C., Chancel, M., & Ehrsson, H. H. (2023). Quantifying body ownership information processing and perceptual bias in the rubber hand illusion. Cognition, 238, Article 105491. https://doi.org/10.1016/j.cognition.2023.105491
- Lenggenhager, B., Tadi, T., Metzinger, T., & Blanke, O. (2007). Video ergo sum: Manipulating bodily self-consciousness. Science, 317(5841), 1096–1099. https://doi.org/10.1126/science.1143439
- Longo, M. R., & Haggard, P. (2010). An implicit body representation underlying human position sense. Proceedings of the National Academy of Sciences, 107(26), 11727–11732. https://doi.org/10.1073/pnas.1003483107
- Longo, M. R., Azañón, E., & Haggard, P. (2010). More than skin deep: Body representation beyond primary somatosensory cortex. Neuropsychologia, 48(3), 655–668. https://doi.org/10.1016/j.neuropsychologia.2009.08.022
- Longo, M. R., & Haggard, P. (2012). What is it like to have a body? Current Directions in Psychological Science, 21(2), 140–145. https://doi.org/10.1177/0963721411434982
- Macaluso, E., & Maravita, A. (2010). The representation of space near the body through touch and vision. Neuropsychologia, 48(3), 782–795. https://doi.org/10.1016/j.neuropsychologia.2009.10.010
- Makin, T. R., Holmes, N. P., & Ehrsson, H. H. (2008). On the other hand: Dummy hands and peripersonal space. Behavioural Brain Research, 191(1), 1–10. https://doi.org/10.1016/j.bbr.2008.02.041
- Makin, T. R., & Flor, H. (2020). Brain (re)organisation following amputation: Implications for phantom limb pain. NeuroImage, 218, Article 116943. https://doi.org/10.1016/j.neuroimage.2020.116943
- Maravita, A., & Iriki, A. (2004). Tools for the body (schema). Trends in Cognitive Sciences, 8(2), 79–86. https://doi.org/10.1016/j.tics.2003.12.008
- Mortensen, E. S., & Christensen, M. S. (2026). Proprioceptive integration in motor control. The Journal of Physiology, 604(8), 3431–3456. https://doi.org/10.1113/JP289835
- Petkova, V. I., & Ehrsson, H. H. (2008). If I were you: Perceptual illusion of body swapping. PLOS ONE, 3(12), Article e3832. https://doi.org/10.1371/journal.pone.0003832
- Proske, U., & Gandevia, S. C. (2012). The proprioceptive senses: Their roles in signaling body shape, body position and movement, and muscle force. Physiological Reviews, 92(4), 1651–1697. https://doi.org/10.1152/physrev.00048.2011
- Rohde, M., Di Luca, M., & Ernst, M. O. (2011). The rubber hand illusion: Feeling of ownership and proprioceptive drift do not go hand in hand. PLOS ONE, 6(6), Article e21659. https://doi.org/10.1371/journal.pone.0021659
- Sainburg, R. L., Ghez, C., & Kalakanis, D. (1999). Intersegmental dynamics are controlled by sequential anticipatory, error correction, and postural mechanisms. Journal of Neurophysiology, 81(3), 1045–1056. https://doi.org/10.1152/jn.1999.81.3.1045
- Scotto, C. R., Ferrier, C., Bosquet, L., & Toussaint, L. (2026). Exercise-induced muscle soreness impairs the efficiency of proprioceptive integration and body representation processing. Journal of Neurophysiology, 135(5), 1211–1219. https://doi.org/10.1152/jn.00004.2026
- Tosi, G., Mentesana, B., & Romano, D. (2023). The correlation between proprioceptive drift and subjective embodiment during the rubber hand illusion: A meta-analytic approach. Quarterly Journal of Experimental Psychology, 76(10), 2197–2207. https://doi.org/10.1177/17470218231156849
- Tosi, G., Kalckert, A., Sivasubramanian, A. K., & Romano, D. (2024). The rubber hand illusion questionnaire: An exploratory graph analysis of ownership, referral of touch, and control statements. Attention, Perception, & Psychophysics, 86, 2866–2876. https://doi.org/10.3758/s13414-024-02964-w
- Tsakiris, M., & Haggard, P. (2005). The rubber hand illusion revisited: Visuotactile integration and self-attribution. Journal of Experimental Psychology: Human Perception and Performance, 31(1), 80–91. https://doi.org/10.1037/0096-1523.31.1.80
- Tsakiris, M. (2010). My body in the brain: A neurocognitive model of body-ownership. Neuropsychologia, 48(3), 703–712. https://doi.org/10.1016/j.neuropsychologia.2009.09.034
- van Beers, R. J., Sittig, A. C., & Denier van der Gon, J. J. (1998). The precision of proprioceptive position sense. Experimental Brain Research, 122(4), 367–377. https://doi.org/10.1007/s002210050525
- van Beers, R. J., Sittig, A. C., & Denier van der Gon, J. J. (1999). Integration of proprioceptive and visual position-information: An experimentally supported model. Journal of Neurophysiology, 81(3), 1355–1364. https://doi.org/10.1152/jn.1999.81.3.1355
- Wolpert, D. M., Ghahramani, Z., & Jordan, M. I. (1995). An internal model for sensorimotor integration. Science, 269(5232), 1880–1882. https://doi.org/10.1126/science.7569931