In the last twenty years a robust experimental paradigm has emerged for studying the structure of bodily experience, focusing primarily on what it is to experience one’s body as one’s own. As in many areas of psychology and cognitive science, induced illusions have proved highly illuminating. The initial impetus came from the rubber hand illusion (RHI) first demonstrated by Botvinick and Cohen (1998), subsequently extended by various researchers to generate illusions of ownership at the level of the body as a whole (for reviews, see Tsakiris 2010; Serino et al. 2013; Kilteni et al. 2015). The resulting experimental paradigms and results have allowed cognitive scientists to operationalize aspects of bodily experience previously explored either purely theoretically or as distorted in neurological disorders such as unilateral spatial neglect. Experiments have focused primarily on the experience of ownership (what it is to experience one’s body as one’s own) and the experience of agency (what it is to experience acting with and through one’s body).
However, these illusion paradigms have not directly studied one important aspect of bodily experience, namely, how the space of the body is experienced. The experienced spatiality of the body is part of what marks out our experience of our bodies as unique physical objects (Bermúdez 1998, 2005, 2011). There is, of course, no objective difference between the space of the body and the space of the extrabodily environment. But the representation of spatial location in bodily experience is very different from the representation of spatial location in vision and other forms of exteroceptive perception. Understanding these differences is integral to understanding how bodily experience contributes to self-consciousness.
This essay identifies some problems with how ownership is discussed in the context of bodily illusions, and then shows how those problems can be addressed through a model of the experienced space of the body. Section 1 briefly reviews the bodily illusions literature and its significance for cognitive science and philosophy. Section 2 expresses reservations with the concept of ownership in terms of which the RHI and other illusions are standardly framed. I offer three hypotheses for the source of our putative sense of ownership. The main body of the essay develops the third hypothesis, which is that judgments of ownership are grounded in the distinctive way that we experience the space of the body.
I
Botvinick and Cohen (1998) introduced an experimental paradigm for inducing illusions of ownership that has been widely replicated and developed. Since 1998, over 200 papers have been published on the RHI, in which subjects see a rubber hand being stroked while their own hand, which they cannot see, is being synchronously stroked. Subjects reliably report both feeling sensations of touch in the rubber hand and feeling that the rubber hand is their own hand. A number of studies have suggested that these subjective reports correlate reliably with behavioral measures, including
galvanic skin conductance response (Armel and Ramachandran 2003),
temperature in the stimulated hand (Mosely et al. 2008)
drift in the perceived location of the hand relative to the rubber hand (Tsakiris and Haggard 2005)
The subjective experience of subjects undergoing the illusion has also been studied psychometrically with a 27-item questionnaire subsequently put through a principal component analysis (Longo et al. 2008).
Further development of the RHI paradigm has explored the possibility of dissociating reports of ownership in the rubber hand and reports of agency with respect to the hand (Kalkert and Ehrsson 2012); the significance of postural and anatomical congruence between the real hand and the rubber hand (Costantini and Haggard 2007); and the possibility of eliciting the illusion without visuotactile stimulation (Kalkert and Ehrsonn 2012). In terms of understanding the mechanisms underlying the RHI, the main issue has been whether the experience of ownership is a bottom-up, stimulus-driven process, or whether it involves higher-order representations of the body, such as a long-term body image and/or an occurrent postural map (Tsakiris 2010; de Vignemont 2014). Investigations of the neural underpinnings of the experience of bodily ownership have highlighted the role of the temporal-parietal junction (Ionta et al. 2014).
Comparable illusions have been induced for the sense of ownership of the entire body using head-mounted displays to create a virtual reality in which subjects see the world from a different spatial perspective. The illusion has been generated from different perspectives, both third person and first person, as illustrated in figure 8.1. In the full-body illusion, subjects can be brought to identify with a full-body avatar that they see being stroked at a spatial location in front of them (Lengenhagger et al. 2007). In the body-swap illusion, subjects reports a sense of ownership for a mannequin that they see being stroked in the location where they would expect their own bodies to be (Petkova and Ehrrson 2008). An interesting phenomenon noticed in studies of the full-body illusion is that the illusion persists when transposed into a third-person perspective—subjects can perceive being in front of their own bodies and even shaking hands with themselves.
Figure 8.1 Illustration of the differences in perspective in the full-body illusion (A) and the body-swap illusion (B). From Serino et al. 2013.
One reason that the RHI and related experimental paradigms have generated such interest and excitement in the cognitive science community is that they offer a way of operationalizing and studying aspects of subjective experience that had stubbornly resisted experimental treatment in normal subjects. Previously, most of what was known about bodily experience came from neuropsychological studies of disorders such as unilaterial spatial neglect (in which neurologically damaged patients neglect one side of their bodies, and indeed one side of peripersonal and external side—typically the side opposite the lesion), somatoparaphrenia (in which patients report disownership of body parts, typically on the contralesional side also) and alien hand syndrome (in which brain-damaged subjects report that someone else is moving their hands).1 Although the study of neuropsychological disorders has been very illuminating, it has inherent disadvantages. Disorders such as somatoparaphrenia and alien hand syndrome are rare. It is hard to compare across cases since there is often significant additional brain damage that makes it difficult to disentangle the relative contributions of different disorders. And the reports of patients with severe brain damage can be challenging to understand and interpret.
From a more philosophical perspective, the structure of bodily experience has been studied by philosophers working within both analytical and phenomenological traditions—in fact, this has been one of the relatively few areas where there has been productive dialogue between the two traditions. Some of this work has been done more or less from the proverbial armchair (e.g., O’Shaughnessy 1980). But others have engaged more directly with the scientific study of the body. Merleau-Ponty, for example, was very well read in the neuropsychology of his day and rested important parts of his analysis of bodily experience in The Phenomenology of Perception (Merleau-Ponty 1968) on the detailed analysis of brain-damaged patients (most famously, the German war veteran Schneider). For more recent examples, see the essays in Bermúdez, Marcel, and Eilan 1995, as well as Gallagher 2005, Bermúdez 1998, 2011, and de Vignemont 2014. Many of these philosophers studying the body have converged on a small set of basic concepts and explanatory tools. These include different ways of thinking about the body image, as well as ideas of bodily ownership and bodily agency. These are precisely the notions that are operationalized and manipulated in the RHI and related bodily illusions.
II
The bodily illusions are standardly discussed as illusions of ownership—limb ownership in the case of the RHI and bodily ownership in the body-swap and full body illusions. It is clear that the RHI manipulates subjects’ sense of their body parts belonging to them. If the rubber hand is experienced as being part of one’s body, then it is natural to describe this as the experience of “owning” the rubber hand. But ownership is a rather tenuous and metaphorical concept in this context. We do not own hands, rubber or otherwise, in the way that we own personal property. And it is unlikely that any way of unpacking the metaphor that will work for limb ownership can be applied straightforwardly to so-called bodily ownership. Intuitively one might say that to feel ownership for a limb is to experience it as connected with the rest of one’s body and as a part of one. But could that be the same sense in which we might feel ownership for our bodies as a whole? What would it mean to say that one experiences one’s body as a part of oneself?2
Certainly, it seems impossible to extend to the body as a whole a mereological conception of ownership, as developed by Martin (1995). Martin defines the sense of ownership as a “phenomenological quality, that the body part appears to be a part of one’s body” (1995, 269). This is perhaps one reason that some authors have proposed an account of ownership in terms of the locus of experience (with the emphasis more on ownership of bodily experiences, rather than on ownership of specific body parts). Gallagher, for example, writes of “the sense that I am the one who is undergoing an experience. For example, the sense that my body is moving regardless of whether the movement is voluntary or involuntary” (2000, 15).
Thinking about ownership in terms of the locus of experience is applicable both to individual body parts and to the body as a whole. However, there is a fundamental lack of clarity in the idea of a sense of ownership as it features in the literature. When, for example, it is said that subjects in the RHI have a sense of ownership of the rubber hand, does this mean that there is a specific feeling of ownership—a qualitative “feel” that one has in all and only those body parts that one experiences as one’s own? This is how some authors have described the phenomenology of ownership (see, e.g., Gallagher 2005 and de Vignemont 2007, 2013). But I have argued elsewhere that the feeling of ownership is a philosophical fiction (Bermúdez 2011, 2015). If there is no such specific feeling of ownership then the putative sense of ownership becomes something that itself needs to be explained, rather than a basic notion that can do explanatory work. We know that there are judgments of ownership (and judgments of disownership, for that matter, as in somatoparaphrenia). If there is a specific feeling of ownership, then those judgments can be viewed as simply reports of the feeling (or perhaps as expressions of the feeling, depending on one’s view of first-person phenomenological reports). But if there is no such feeling of ownership then some account needs to be given of what those judgments are based on. This will, in effect, be a substantive account of what the sense of ownership consists in.3
So, when subjects report ownership (or disownership) of limbs, or of their entire bodies, what are those judgments based on? Here are three hypotheses:
(1) Judgments of ownership are based on the experienced location of sensation.
(2) Judgments of ownership are based on the experience of agency.
(3) Judgments of ownership are based on the experienced spatiality of the body.
This section explains why (1) and (2) are unconvincing. The remainder of the essay will focus on (3).4
The idea behind (1) is that we report ownership of those limbs and body parts where we experience sensations. This idea goes back (at least) to the seventeenth century. In his Essay Concerning Human Understanding, John Locke argues that our “thinking, conscious self” extends to all parts of the body “that we feel when they are touch’d, and are affected by, and conscious of good or harm that happens to them” (Locke 1689/1975, Bk. II, chap. 27, §11). On this view, the body is experienced as the locus of sensation, so that judgments of ownership follow the localization of sensations.5 One aspect of Locke’s account that needs to be brought out is his stress on the dimension of concern. For Locke, to experience individual body parts and one’s body as a whole as parts of one’s “thinking, conscious self” is ipso facto to feel an immediate concern for what happens to them. This affective aspect of ownership has been highlighted by de Vignemont (forthcoming).
However, experiencing sensations in a particular body part appears to be neither necessary nor sufficient for judgments of ownership. Somatoparaphrenia seems to show that it is not sufficient. Somatoparaphrenic patients deny ownership of a limb or even of an entire side of their body. Yet there are reports from somatoparaphrenic patients of feeling sensations in limbs that they deny are theirs (Aglioti, Smania, Manfredi, and Berlucchi 1996; Bottini, Bisiach, Sterzi, and Vallar 2002).6 And there are examples from the bodily illusion literature showing that experiencing sensations in a particular body part is not required for judgments of ownership. The experiments described by Ferri et al. (2013) show that the illusion can be induced by subjects seeing a rubber hand being approached, even when there is no touch (either of the rubber hand or of their own hand).
The idea that there is a connection between ownership and agency is highly intuitive, so that we experience our body and body parts as our own to the extent that we act with them, or are capable of acting with them. Within the philosophical literature, this way of thinking about judgments of ownership has typically not been distinguished from the first hypothesis. There are some good reasons for this, as will emerge in the next three paragraphs—and also some more questionable ones, connected to the postulation of a sense of agency but also capable of being dissociated from the sense of ownership (see Gallagher 2005 for an influential formulation). To my mind, the putative sense of agency suffers from exactly the same explanatory problems as the putative sense of ownership. In any event, setting up a dialectic on which there are distinct and separable sense of ownership and agency obscures the potential role of awareness of agency in underwriting judgments of ownership.7
Whereas hypothesis (1) frames the body as the locus of sensation, hypothesis (2) frames the body as the locus of action. Nonetheless, the rubber hand illusion itself shows that judgments of ownership cannot be based solely on the experience of acting with specific body parts, since there is no active movement in the RHI. The experience of agency cannot be a necessary condition. It may be a sufficient condition, however. Kalkert and Ehrsonn (2012) showed that the RHI can be induced with active movement in the absence of sensory stimulation.
Still, hypothesis (2) seems too narrowly formulated. For one thing, focusing exclusively on the experience of agency is insufficiently distant from hypothesis (1), since a significant part of the experience of agency surely consists in kinesthetic and other bodily sensations. An alternative would be to take judgments of ownership to rest on the capacity for agency, rather than the experience of agency. The guiding idea here is that the body is the physical object uniquely responsive to the will, and so we take ownership to extend to body parts that we can act with directly.
One difficulty with this proposal is that there are body parts that we judge to be our own but that we act with only in a very attenuated sense (if at all), such as the spleen or the kidney. Another, perhaps more significant, challenge is operationalizing the capacity for agency, as opposed to the exercise of agency in action. What would be an adequate test of whether one experiences a body part as something that one can act with, in circumstances in which one does not actually move it? The problem is that it is not clear what would distinguish the judgment that I can act with this limb from the corresponding judgment of ownership (this is my limb)—and yet we need the former judgment to be the sort of thing that can ground the latter judgment.
A more promising approach, I think, would be to look for a phenomenon that might plausibly explain why both the experienced location of sensation and the experience of agency seem so important to judgments of ownership. This is where hypothesis (3) comes into the picture. Both forms of experience reflect the fact that we experience our bodies as distinctive physical objects. They also reflect (and possibly even contribute to) a very important source for that experienced distinctiveness, namely, the differences between how space is represented within the bounds of the body and how extrabodily space is represented—or, to put it another way, between how bodily locations are encoded and how nonbodily locations are encoded. So, the hypothesis that I will explore in this essay is that judgments of ownership track the distinctive spatiality of bodily experience.
III
This section focuses on two very general features of how bodily events are experienced. (I am understanding bodily events in a broad sense, to include sensations such as itches, pains, and so forth, as well as tactile experiences and proprioceptive/kinesthetic experiences of how one’s limbs are disposed and/or moving.) The first feature I call Boundedness.
Boundedness
Bodily events are experienced within the experienced body (a circumscribed body-shaped volume whose boundaries define the limits of the self).8
The body-shaped circumscribed volume that defines the limits of the self is the experienced body. It can be, but need not be, identical to the physical body. There are many documented cases where the two diverge, and the divergence takes place in both directions. The boundaries of the bodily self in the experienced body can extend beyond the limits of the physical body. The experience of sensations in phantom limbs is a well-known example, indicating how the experienced bodily self expands to accommodate displaced sensation. Pathologies such as unilateral spatial neglect and somatoparaphrenia illustrate how the experienced body can be more restricted than the bounds of the physical body.
I should clarify how Boundedness is to be understood. It is not intended as a necessary truth about human somatic experience. There seem to be counterexamples in the literature. I am grateful to a referee for directing me to an interesting case presented by Cronholm (1951). Cronholm reports a phantom limb patient at the Carolinska Institut in Stockholm who appeared to be fully aware of the spatial boundaries of his phantom and yet who reported feeling a sensation in a region of space outside those boundaries. This case is highly unusual, but I would be surprised if it were unique. From my perspective, however, Boundedness is proposed as a general characterization of the phenomenology of bodily awareness. That it might be contravened in highly unusual circumstances is interesting, but not particularly damaging to the proposal, particularly given that Boundedness does seem to be respected in the overwhelming majority of pathological cases.
In any event, Boundedness captures a reciprocal, temporally extended, and plastic process. At any given moment the boundaries of the experienced body are relatively fixed. But viewed over time the experienced body is malleable and adaptable, responding to organic bodily growth, trauma, and the changing demands of movement and action. The technique of using extended physiological proprioception (EPP) in the design of prosthetics for amputees is an excellent illustration of the role that agency can play in redefining the limits of the experienced self. The inspiration for EPP comes from the familiar example of blind people using white canes not just as motility aids but also as tools for discovering the environment. Recognizing this type of transferred sensation as a potential tool for improving the effectiveness of prosthetic limbs, D. C. Simpson proposed EPP in the 1970s.
The basic idea of EPP is to design prostheses so that feedback information from the artificial limb is experienced proprioceptively. Agency is crucial, since the mechanism for achieving this is coupling the movement of the prosthetic with the residual movements available through the nearest intact joint so that there is a direct relation between the movement and position of the prosthetic limb and the movement and position of the anatomical joint. We can think about the movement of a prosthetic limb in input–output terms. The movement of the intact joint is the input and the new position of the prosthetic is the output. By designing the prosthetic limb so that the input and output are closely bound to each other, the output position can be sensed through proprioceptive feedback from the input movement. It appears that using the EPP technique to design prosthetic limbs improves their effectiveness (Doubler and Childress 1984).
There are no studies that I am aware of on the subjective experience of amputees with EPP, but we can obtain guidance from qualitative studies of patients with “ordinary” prosthetics. A study of thirty-five amputees in the UK specifically explored the experience of embodiment in wearers/users of prosthetics, with many of the patients reporting their phantom limbs merging with the prosthesis (Murray 2004). Here are some sample comments from amputees:
… because I don’t feel as anything is really missing. So my prosthesis is “natural.” (Murray 2004, 969)
When I put on a prosthetic, the phantom becomes the prosthesis to the extent that the not-foot [phantom] is in almost the same position as the Flexfoot [a brand of prosthesis], maybe slightly more rotated. The fit is so good, that it makes walking with the prosthesis easier because of the correspondence between the prosthetic leg and the phantom. (970)
… Many amputees feel that their artificial limb is somehow part of them, a simple example of this is that I wouldn’t like just anyone putting their hand on my artificial knee, even though it is not actually part of my body’s flesh, it is still mine even though it’s a piece of plastic and metal. (970)
Interviewer
When you say it’s part of you now, what exactly do you mean by that?
Participant
Well, to me it’s as if, though I’ve not got my lower arm, it’s as though I’ve got it and it’s [the prosthesis is] part of me now. It’s as though I’ve got two hands, two arms. (970)
This “merging” of phantom limbs with prostheses contrasts significantly with referred sensations in tool use and with the incorporation of tools in the body schema more generally. Here the prosthetic seems to be incorporated into the limits of the self, in the way that tools typically are not.9
The second general feature of how bodily events are experienced I term Connectedness.
Connectedness
The spatial location of a bodily event is experienced relative to the disposition of the body as a whole.
Connectedness presupposes Boundedness, but not vice versa. It is conceptually possible that we could experience bodily events within a space that defines the limits of the self and yet experience those events in isolation from everything else going on in that space. But that is not the normal phenomenology of the body. Bodily events are typically experienced relative to the background of the body as whole. In experiencing a pain in my knee, for example, I experience the pain as being in my leg, which in turn is disposed in a certain way relative to my torso. My torso itself is experienced as being disposed in a certain way relative to gravity and supporting surfaces. In this sense, therefore, bodily events are experienced within a holistic framework that, although sometimes recessive, is normally an ineliminable part of the content of experience. To motivate that thought, think how strange it would be to feel a sensation in your foot without having an idea of the angle of your foot relative to your lower leg, or of whether your leg is bent or straight at the knee.
The phenomena of both Boundedness and Connectedness can be readily identified in the bodily illusion experiments. So, for example, Boundedness predicts that subjects experience the rubber hand as incorporated into their own bodies—as opposed, for example, to being spatially discontinuous with their bodies. This is exactly how the phenomenology of the RHI is described by the subjects reviewed by Longo et al. (2008). Moreover, as Longo et al. observe, the rubber hand is typically experienced as displacing the subject’s real hand. It is very rare for subjects in RHI experiments to report experiencing the rubber hand as a third, supernumerary hand (although see Ehrsson 2009 for a version of the RHI in which healthy subjects seem to report experiencing two right hands). Interestingly, in the so-called invisible hand illusion, where referred sensations are elicited by stroking empty space, subjects report feeling the referred sensations in a hand that they cannot see, rather than outside the body (Gutersdam, Gentile, and Ehrsson 2013).
As this indicates, bodily awareness can only be manipulated within certain, structural limits. One limiting factor is what is sometimes called the long-term body-image (O’Shaughnessy 1980, 1995)—an implicit understanding of the large-scale, structural properties of the body (such as the property of having no more than two hands!). A number of studies have shown that the RHI is constrained by factors of anatomical plausibility. The illusion is extinguished, for example, when the rubber hand is of a different laterality from the subject’s own hand (Tsakiris and Haggard 2005). This is another Boundedness effect.
Manipulations of the RHI paradigm also reveal Connectedness effects. Most significantly, postural mismatches can extinguish the illusion. The difference between a postural mismatch and an anatomical mismatch is important. In both cases the manipulation extinguishes the illusion by, in effect, revealing it to be impossible for the subject to be the owner of the rubber hand. This comes about because the rubber hand is placed in a position that conflicts with subjects’ knowledge of their own bodies and how they are disposed. As we have seen the knowledge engaged in Boundedness effects is general knowledge of the structure of the body. In cases of postural mismatch, in contrast, the rubber hand is (as it were) anatomically viable. But it just couldn’t be part of the subject’s body because the disposition of the rubber hand is inconsistent with the subject’s knowledge of how his actual hand and arm are disposed. Tsakiris and Haggard (2005), for example, established that the illusion disappears if the rubber hand is oriented at a 90° angle to the subject’s actual hand. The angle of the rubber hand rules out its being attached to the subject’s wrist and arm. This is a Connectedness effect, because if the position of the hand were not coded relative to the position of the wrist and so on there would be no inconsistency.
There are clear connections between, on the one hand, Boundedness and Connectedness and, on the other, the experienced location of sensation (hypothesis 1) and the experience of agency (hypothesis 2). One reason the experienced location of sensations seems so important for ownership is that (per Boundedness) sensations are typically experienced within the limits of the bodily self. Likewise for the experience of agency, which is standardly enabled by the experience of a connected and bounded body.10 Boundedness and Connectedness are more general aspects of the phenomenology of bodily awareness, however. Letting them carry the explanatory weight allows us to think of judgments of ownership as multifactorial. The experience of agency and the experienced location of sensations both play a role in grounding judgments of agency. In the next section, we will see that both Boundedness and Connectedness are themselves ultimately grounded in certain very basic features of the spatiality of bodily experience.
IV
How is the space of the body represented? The general topic of somatic spatial representation has seen rapidly increasing attention from psychologists and neuroscientists over the last three decades. Researchers have pursued a number of different, but overlapping, questions highly germane to the discussion in earlier sections of bodily illusions and the phenomenology of bodily awareness, but the two lines of research do not always map cleanly and directly on to each other. The principal focus of research into bodily illusions is on how we experience our bodies as our own. The principal focus of research into the neuroscience of spatial representation has been on how the brain encodes information about the body and about the location of objects in the immediate distal environment to allow reaching and other motor behaviors. Of course, it is hard to imagine that the way we experience our bodies is not at least partially determined by how the brain encodes spatial information, but the fact remains that these are two different questions. This section disentangles some of these issues.
One much discussed issue is how we represent peripersonal space. Peripersonal space is standardly defined as the area of space around the body that is within reach—as opposed to extrapersonal space, which is out of reach. Whereas information about extrapersonal space comes primarily through vision, with contributions from hearing and smell, awareness of peripersonal space is much more multimodal. An elegant illustration of the difference between peripersonal and extrapersonal space comes from experiments on patients with unilateral spatial neglect. Neglect patients typically make significant errors when asked to bisect lines—since they neglect (typically) the left side of space, they will place the midpoint of the line far to the right of the true midpoint. Since these bisection tasks involve drawing a line they take place within peripersonal space. However, Halligan and Driver (1991) showed that a neglect patient with right hemisphere damage after a stroke, despite having the standard deficit when asked to draw a bisecting line, showed no deficit when asked to use a laser pointer to bisect a line in extrapersonal space.
There is considerable consensus that the representation of peripersonal space engages multiple frames of reference, depending on the relevant modality (Spence and Holmes 2004; Battaglia-Meyer et al. 2003). Information from vision is standardly represented in retinotopic coordinates (coordinates centered on the retina), while auditory and olfactory information is coded in head-centered coordinates (Cohen and Anderson 2002). Moreover, neuron recordings in macaque monkeys have identified arm-centered receptive fields in the premotor cortex (Graziano, Yap, and Gross 1994). These receptive fields move when the arm moves, rather than when the eyes move. So, the representation of peripersonal space is multilevel and varies according to modality and context.
Moreover, Noel et al. (2015) have shown that in the full-body illusion peripersonal space shifts to being centered at the location of the virtual body. However, one cannot draw direct conclusions about how the spatiality of the body is experienced from representations of peripersonal space. It is true that if a receptive field moves with the location of the arm, then the nervous system must have some way of keeping track of where the arm is. But it is perfectly possible, given the design of the studies, that the location of the arm is tracked visually, relative to a retinotopic reference frame. This would tell us nothing about how the body is experienced “from the inside” through bodily sensations, proprioception, and kinesthesis. And even if the location of the arm were tracked “from the inside,” the question is still wide open as to the frame of reference relative to which this tracking takes place.
The question of how somatosensory spatial information is integrated with visual information has been prominent in studies of reaching. Successful reaching depends on calibrating the represented location of the target with the represented starting point of the hand, to allow both the initial aiming and online monitoring and correction of the movement. There is evidence from psychophysics that successful reaching depends on remapping the target location from a retinotopic coordinate system into body-centered and/or hand-centered frames of reference (Soechting and Flanders 1989a, 1989b). Such remapping would plausibly make the calibration easier. But, as with the representation of peripersonal space, it does not really help with the question we are interested in. Exactly the same question arises. To say that a target location is relative to a hand-centered frame of reference just means that target location is coded on a coordinate system whose origin is some designated point in the hand. But that doesn’t tell us how the nervous system encodes the location of the center of the coordinate system—let alone how the location of the hand is experienced “from the inside” relative to the rest of the body.
What is needed to complement ongoing research into how peripersonal space and reaching movements are neutrally encoded is a model of how normal subjects experience the space of the body. Such a model obviously needs to be consistent with the existing experimental literature, but also needs to go further in two key respects. First, it needs to focus on the phenomenological aspect of spatial awareness of the body. That is, it needs to focus on how subjects are conscious of bodily space—on what it is to experience a bodily sensation at a particular bodily location, for example, or how the layout of the body is presented in somatosensory experience. Second, it needs to explain why our experience of our own bodies has the properties of Boundedness and Connectedness.
V
The first issue that arises in modeling the spatial content of bodily awareness is determining a frame of reference and corresponding coordinate system. This section introduces two possible approaches.
The first approach is in essence a direct continuation of the modeling strategies discussed in the previous section in the context of peripersonal space and reaching. We looked at a number of different reference frames—retinotopic, head-centered, or centered on specific body parts such as the hand. These are all standard Cartesian frames of reference with three axes. What distinguishes them is the point each takes as its origin. A natural extension of this approach would be to conceptualize the space of the body in a Cartesian frame with three axes, corresponding to the frontal, saggital, and transverse planes. A plausible candidate for the origin of the coordinate system would be the body’s center of mass.
So, on this modeling strategy, to experience a bodily event at a particular bodily location is to experience it at a certain point (x, y, z) in a space whose origin is the body’s center of mass, where that point is given in terms of its distance from the origin on each of the three axes. A variant (probably more plausible) would be to use a spherical coordinate system, generalizing two-dimensional polar coordinates, rather than Cartesian coordinates, so that the sensation is given as located at a point (r, θ, φ) where r is the radial distance from body’s center of mass; θ is the polar angle (measured from a fixed zenith in the vertical direction opposite the pull of gravity); and φ the azimuth angle (measured from a fixed direction on a plane orthogonal to the zenith, and so at right angles to the pull of gravity). Cartesian and spherical coordinate systems are interconvertible, of course.
One advantage of this approach is that it allows the space of the body to be mapped straightforwardly onto peripersonal space. A target location perceived in hand-centered coordinates in peripersonal space can equally be described in Cartesian or spherical coordinates relative to an origin in the body’s center of mass. And so, for example, it is computationally easy to plot the displacement required to move the hand from its current location to the target location, as well as to monitor the movement and adjust while it is in progress.
On the other hand, though, the approach has difficulty doing justice to how we actually experience our bodies. For one thing, as I have pointed out in previous work, our experience of our own bodies does not typically present a particular point as a privileged origin so that we experience particular bodily locations as being nearer than or further away from that origin (Bermúdez 1998, 2005, 2011). In fact, it is not clear that the concepts nearer than or farther away from have any applicability to the space of bodily experience. These comparative concepts have an implicit self-reference built into them, unless some other object is explicitly given as a reference point. To say that one thing is nearer than another is typically to say that it is nearer to me. Within a visual perspective this makes perfectly good sense, since the structure of the visual field implicitly defines a point that can serve as me for the purposes of comparison. But bodily awareness is not like this. The body’s center of mass is important, of course, for balance, and so more generally for movement, but not in a way that makes it a candidate for calibrating distance and direction. And there is no other point in the body that can count as me relative to the rest of the body in the same way that the origin of the visual field counts as me relative to objects in the distal environment. So comparisons of distance and direction really only make sense in special circumstances (talking to the doctor, for example, where a reference point is typically assumed).
A second problem with the standard approach comes with accommodating Boundedness and Connectedness. According to Boundedness, bodily events are experienced within the experienced body (a circumscribed body-shaped volume whose boundaries define the limits of the self). So, from the perspective of conscious bodily experience, there is a marked distinction between bodily space and peripersonal space. Within the total volume of space defined by the limits of peripersonal space, those points that fall within the perceived space of the body are experientially privileged. They are, as Locke would put it, part of the conscious self. Yet this distinction is not at all captured within the model under consideration. Within the context of a frame of reference centered on the body’s center of mass, the space defined is completely homogeneous. There is no distinction between a point that falls within the perceived space of the body and one that does not—between a point that falls within my left forearm, for example, and a point that is three inches to the right in extra personal space. The standard approach would admittedly allow bodily boundaries to be defined, but this would be (as it were) a purely geometrical way of marking the difference between bodily space and peripersonal space. It would not in any sense capture how the space of the body is experientially privileged.
The standard approach fares no better with Connectedness. According to Connectedness, the spatial location of a bodily event is experienced relative to the disposition of the body as a whole. It follows that experienced locations within bodily space are not experienced in isolation. So, when we experience our foot flexed at a certain angle we experience that flexing of the ankle relative to the disposition of the foot, the knee, and the hip, for example. That is to say, bodily experience is both relational and holistic. In contrast, spatial locations on the standard approach are purely particularist. A bodily location is given as a triple of numbers, either (x, y, z) or (r, θ, φ). The coding of the location carries no information about what is going on elsewhere in the body.
For these reasons I have proposed a different model for thinking about how the space of the body is represented in conscious bodily experience. A starting point for the model is a distinction between two ways of thinking about bodily location. We can think about a given bodily location in a way that abstracts from the disposition at a time of the body as a whole. So, for example, we might think about an itch being located at the front of the shin. This is a location that the itch has as long as it endures, irrespective of whether or not the shin moves. Alternatively, one might think of the itch in a way that takes into account what the rest of the body is doing. In previous work I have termed these A-location and B-location, respectively (Bermúdez 1998, 2005, 2011).
A-location (1)
The location of a bodily event in a specific body-part relative to an abstract map of the body, without taking into account the current position of the body.
B-location (1)
The location of a bodily event in a specific body-part relative to the current position of relevant body-parts.
So, for example, if I have a pain in the middle finger of my left hand and then contract my bicep to raise my left hand by six inches, then the A-location of my pain remains unchanged, while the B-location changes. The A-location remains unchanged because the pain is still in the middle finger of my left hand. The B-location is different, however, because my left hand is now at a different angle relative to my elbow. Both A-location and B-location are body relative. If I move six feet to the left and then stand as I am standing here, the pain in the middle finger of my left hand will still have the same A-location and B-location.
The idea, therefore, is that we experience a given bodily event at a specific A-location and B-location. A-location and B-location are not independent of each other—two different aspects of a single type of experience rather than two different ways of experiencing bodily events. The A-location dimension of bodily experience does justice to Boundedness, while the B-location dimension speaks more to Connectedness. If we typically experience a bodily event at a specific A-location then it follows automatically that we experience it within the experienced body, because A-locations can only fall within the perceived limits of the body. By the same token, if we typically experience a bodily event at a specific B-location, then we gain immediate insight into the relation between that bodily event and the overall disposition of the body.11
Of course, the explanatory value of this model depends on giving a substantive account of A-location and B-location. In particular, we need an account of how the body is represented in somatosensory experience that will explain how and why we experience bodily events at specific A-locations and A-locations. I offer such an account in the next and final section.
VI
Extending Bermúdez 1998 and subsequent papers (2005, 2011), the previous section offered a general account of how the space of the body is represented in experience. This section offers a more detailed account that develops insights from two approaches to modeling the body—in biomechanics and robotics, on the one hand, and in Marr and Nishihara’s model of object recognition, on the other.
The model proposed in The Paradox of Self-Consciousness (Bermúdez 1998) starts from how the body is articulated as a relatively immoveable torso connected by joints to moveable body parts. These joints range in size and scope from the knee and the neck, at one extreme, to the joints in fingers and toes, at the other. Joints afford the possibility of moving the body parts that they connect. There are around 230 joints in the human body, differing in the types of movement that they allow and in the degrees of freedom that they offer. The hip, for example, operates on all three planes and allows six different types of movement (abduction, adduction, extension, flexion, horizontal abduction, and horizontal adduction). The knee, like many of the finger joints, allows only flexion and extension in a single plane.
Joints provide the fixed points relative to which particular A-locations and B-locations can be specified.
A-location (2)
A particular bodily A-location is experienced in a given body part and specified relative to the joint immediately controlling the movement of that body part.
So, for example, an itch in the palm of my hand is experienced in the palm of my hand and its location is specified relative to the wrist, since the wrist immediately controls the movement of the palm of my hand (as opposed to the elbow and shoulder, whose control is mediate rather than immediate).
B-location (2)
A particular B-location is a given A-location oriented in a certain way relative to the rest of the body. It is specified recursively relative to the joints that lie between it and the immoveable torso.
The B-location of the itch in the palm of my hand is its A-location, supplemented by specifying the angles of the wrist relative to the forearm, the elbow relative to the upper arm, and the shoulder relative to the torso.12
To flesh this general model out further, we can draw on two different but complementary approaches to modeling the body. The first comes from biomechanics and robotics, where the body is typically modeled as a system of rigid links connected by mechanical joints. Kinesiologists and roboticists have developed sophisticated tools for representing both bodily position at a time and movement over time in terms of the angles of the relevant joints (see, e.g., a standard biomechanics textbook such as Hall 2014).
The second derives from Marr and Nishihara’s model of object recognition in terms of matching three-dimensional object representations derived from vision with a stored catalog of three-dimensional representations (Marr and Nishihara 1978). Objects are represented schematically as complexes of generalized cones (surfaces generated by moving a cross-section along an axis, maintaining its shape but possibly varying its size). The human body is represented as a hierarchy of generalized cones, as depicted in figure 8.2.
Figure 8.2 Marr and Nishihara’s model of the human body as a hierarchy of generalized cones. From Marr and Nishihara 1978.
The suggestion I will develop is that within bodily experience the human body is represented in terms of generalized cones. This approach has two very significant advantages.
First, it allows us to fill in a gap in the general models of the body in robotics and biomechanics. Understandably, given what they are trying to achieve, kinesiologists and roboticists typically abstract away from the details of what goes on inside the limbs. From the perspective of studying human movement or programming artificial movement, it makes perfectly good sense to treat limbs as rigid links with no intrinsic properties. When studying the experienced space of the body, however, we obviously need tools for describing the location of bodily events that take place inside limbs. Modeling limbs as generalized cones allows us to do this in a way that is faithful to the anatomy of the body.
Second, if we think of limbs as generalized cones then we can use a cylindrical coordinate system, somewhat similar to the spherical coordinate system discussed earlier. Identifying a cylindrical coordinate system involves identifying an origin at the intersection of the longitudinal axis (along the cylinder’s axis of symmetry) and a reference plane orthogonal to the longitudinal axis, with a specified reference direction. The location of a point x in the cylinder is given by the triple (ρ, φ, z) where (ρ, φ) are in effect the polar coordinates of x on the plane that passes through x and is parallel to the reference plane and z is the distance from the origin to x along the longitudinal axis. As mentioned earlier, this is just one example of a number of interconvertible coordinate systems.
So, my proposal is that the space of the body is experienced relative to a model of body as a hierarchy of generalized cones linked by mechanical joints. Specifically, the hypotheses that I am making for A-location and B-location are as follows. First,
A-location (3)
The A-location of a bodily event in a given limb is fixed in a cylindrical coordinate system whose origin is at the midpoint (center of mass) of the joint immediately controlling the position of that limb.
The subject’s awareness of the large-scale structure of her body (e.g., what O’Shaughnessy 1980 termed the long-term body image) in effect fixes the surface and volume of the cylinder, thus defining the potential bodily space within which bodily events can be experienced. In most normal cases, of course, there will be a close map between the physical boundaries of the body and the cylinder thus defined. But it is perfectly conceivable that a cylinder should be defined with its origin at the tip of an amputation. This seems to be what is happening when amputees feel sensations in prosthetic limbs, as discussed earlier in the context of extended physiological proprioception.
Second,
B-location (3)
The B-location of a bodily event in a given limb is fixed by supplementing its A-location with a chain of relative joint angles that collectively specify the location of the limb relative to the immoveable torso.
I am using the term “joint angle” as shorthand for whatever information is required to specify the disposition of a joint. It is likely, of course, to involve more than one coordinate, but no joint requires more than six (corresponding to the maximum number of degrees of freedom—three rotational and three translational). The degree of complexity will increase, however, when chains of joint angles are considered.
VII
To recap, experiments on bodily illusions raise important questions about what it is to experience one’s body as one’s own. Philosophers and cognitive scientists often refer to a sense of ownership that extends both to individual body parts and to the body as a whole. I do not find that this approach has much explanatory power, and thus proposed investigating the complex phenomenon of ownership by looking at the grounds for judgments of ownership. After discounting the idea that there is a specific feeling of ownership, three hypotheses emerged.
(1) Judgments of ownership are based on the experienced location of sensation.
(2) Judgments of ownership are based on the experience of agency.
(3) Judgments of ownership are based on the experienced spatiality of the body.
Hypotheses (1) and (2) give an important part of the overall picture, but I argued that neither is fully explanatory on its own. Hypothesis (3) identifies a phenomenon that is more fundamental, since both experiencing agency and experiencing sensations take place relative to experiencing the distinctive space of the body.
In developing hypothesis (3), I began with two large-scale features of the phenomenology of bodily awareness. According to Boundedness, bodily events are experienced within a circumscribed body-shaped volume whose boundaries define the limits of the self, while according to Connectedness, the spatial location of a bodily event is experienced relative to the disposition of the body as a whole. Developing an account of the spatiality of bodily experience that does justice to Boundedness and Connectedness took us to the idea that bodily events are experiences in terms of A-location and B-location, which I then glossed through combining biomechanical models of the body as rigid links connected by mechanical joints with the picture of the body loosely derived from Marr and Nishihara’s model of object recognition. The space of the body is experienced, I suggest, relative to a model of the body as a hierarchy of generalized cones linked by mechanical joints.
Circling back, therefore, to the starting point of the essay, my main proposal is that the judgments of ownership that we make about our bodies and body parts are ultimately grounded in the fact that we experience the space of the body in this distinctive way that differs fundamentally from how we experience extrabodily events, even those in peripersonal space. With this proposal in view, let me end by identifying some important open questions for future work.
How should the account of spatial content proposed in terms of A-location and B-location be developed into a full account of the content of bodily experiences?
How should we spell out in detail the inferential or quasi-inferential relations in which bodily experiences stand to judgments of agency?
How do those inferential or quasi-inferential relations compare to the relations between “ordinary” perceptual experiences and perceptual judgments?
How should the account be extended to cases of disownership, as occur for example in somatopraphrenia and other disorders?
Can a parallel explanatory strategy to that proposed here for the “sense of ownership” be extended to the putative “sense of agency”?
If so, what role does the distinctive spatiality of bodily experience play in underwriting judgments of agency?
Acknowledgments
I am very grateful to Frédérique de Vignemont and Adrian Alsmith and to two anonymous referees for probing and helpful comments; to Patrick Haggard and Tony Cheng for their commentary at the December 2015 Copenhagen workshop; and to the other participants at the Copenhagen workshop.
Notes
References
Aglioti, S., N. Smania, M. Manfredi, and G. Berlucchi. 1996. Disownership of left hand and objects related to it in a patient with right brain damage. NeuroReport 8 (1): 293–296.
Alsmith, A. 2015. Mental activity and the sense of ownership. Review of Philosophy and Psychology 6 (4): 881–896.
Armel, K. C., and V. S. Ramachandran. 2003. Projecting sensations to external objects: Evidence from skin conductance response. Proceedings of the Royal Society of London, Series B: Biological Sciences 270 (1523): 1499–1506.
Battaglia-Mayer, A., R. Caminiti, F. Lacquaniti, and M. Zago. 2003. Multiple levels of representation of reaching in the parieto-frontal network. Cerebral Cortex 13 (10): 1009–1022.
Bermúdez, J. L. 1998. The Paradox of Self-Consciousness. Cambridge, MA: MIT Press.
Bermúdez, J. L. 2005. The phenomenology of bodily awareness. In Phenomenology and Philosophy of Mind, ed. D. W. Smith and A. L. Thomasson, 295–316. New York: Oxford University Press.
Bermúdez, J. L. 2011. Bodily awareness and self-consciousness. In Oxford Handbook of the Self, ed. S. Gallagher, 157–179. Oxford: Oxford University Press.
Bermúdez, J. L. 2015. Bodily ownership, bodily awareness and knowledge without observation. Analysis 75 (1): 37–45. doi:10.1093/analys/anu119.
Bermúdez, J. L., A. J. Marcel, and N. Eilan, eds. 1995. The Body and the Self. Cambridge, MA: MIT Press.
Bottini, G., E. Bisiach, R. Sterzi, and G. Vallar. 2002. Feeling touches in someone else’s hand. NeuroReport 13 (2): 249–252.
Botvinick, M., and J. Cohen. 1998. Rubber hands “feel” touch that eyes see. Nature 391:756.
Cassam, Q. 1995. Introspection and bodily self-ascription. In The Body and the Self, ed. J. L. Bermúdez, A. J. Marcel, and N. M. Eilan, 311–336. Cambridge, MA: MIT Press.
Cassam, Q. 1997. Self and World. Oxford: Oxford University Press.
Cohen, Y. E., and R. A. Andersen. 2002. A common reference frame for movement plans in the posterior parietal cortex. Nature Reviews Neuroscience 3:553–562.
Corbetta, M., and G. Shulman. 2011. Spatial neglect and attention networks. Annual Review of Neuroscience 34:569–599.
Costantini, M., and P. Haggard. 2007. The rubber hand illusion: Sensitivity and reference frame for body ownership. Consciousness and Cognition 16 (2): 229–240.
Cronholm, B. 1951. Phantom limbs in amputees: A study of changes in the integration of centripetal impulses with special reference to referred sensations. Acta Psychiatrica et Neurologica Scandinavica 72 (Suppl.): 1–310.
de Vignemont, F. 2007. Habeas corpus: The sense of ownership of one’s own body. Mind & Language 22:427–449.
de Vignemont, F. 2013. The mark of bodily ownership. Analysis 73:643–651.
de Vignemont, F. 2014. A multimodal conception of bodily awareness. Mind 123 (492): 989–1020.
de Vignemont, F. In press. Mind the Body. Oxford: Oxford University Press.
de Vignemont, F., and A. Farnè. 2010. Widening the body to rubber hands and tools: What’s the difference? (Incorporer des objets et des membres factices: Quelle différence?) Revue de Neuropsychologie 2 (3): 203–211.
Doubler, J. A., and D. S. Childress. 1984. An analysis of extended physiological proprioception as a prosthesis-control technique. Journal of Rehabilitation Research and Development 21 (1): 5–18.
Ehrsson, H. H. 2009. How many arms make a pair? Perceptual illusion of having an additional limb. Perception 38 (2): 310–312. doi:10.1068/p6304.
Ferri, F., A. M. Chiarelli, A. Merla, V. Gallese, and M. Costantini. 2013. The body beyond the body: Expectation of a sensory event is enough to induce ownership over a fake hand. Proceedings of the Royal Society of London, Series B: Biological Sciences 280 (1765): 20131140.
Gallagher, S. 2000. Philosophical conceptions of the self: Implications for cognitive science. Trends in Cognitive Sciences 4 (1): 14–21.
Gallagher, S. 2005. How the Body Shapes the Mind. New York: Oxford University Press.
Gentile, G., M. Björnsdotter, V. I. Petkova, Z. Abdulkarim, and H. H. Ehrsson. 2015. Patterns of neural activity in the human ventral premotor cortex reflect a whole-body multisensory percept. NeuroImage 109:328–340.
Graziano, M. S. A., G. S. Yap, and C. G. Gross. 1994. Coding of visual space by premotor neurons. Science 266 (5187): 1054–1057.
Gutersdam, A., G. Gentile, and H. H. Ehrsson. 2013. The invisible hand illusion: Multisensory integration leads to the embodiment of a discrete volume of empty space. Journal of Cognitive Neuroscience 25 (7): 1078–1099.
Hall, S. 2014. Basic Biomechanics. 7th ed. New York: McGraw Hill.
Halligan, P. W., and J. C. Marshall. 1991. Left neglect for near but not far space in man. Nature 350 (6318): 498–500.
Hamilton, A. 2013. The Self in Question: Memory, the Body, and Self-Consciousness. Basingstoke: Palgrave Macmillan.
Holmes, N. P., and C. Spence. 2004. The body schema and the multisensory representation(s) of peripersonal space. Cognitive Processing 5 (2): 94–105.
Ionta, S., R. Martuzzi, R. Salomon, and O. Blanke. 2014. The brain network reflecting bodily self-consciousness: A functional connectivity study. Social Cognitive and Affective Neuroscience 9 (12): 1904–1913.
Kalckert, A., and H. H. Ehrsson. 2012. Moving a rubber hand that feels like your own: A dissociation of ownership and agency. Frontiers in Human Neuroscience 6: 40.
Kilteni, K., A. Maselli, K. P. Kording, and M. Slater. 2015. Over my fake body: Body ownership illusions for studying the multisensory basis of own-body perception. Frontiers in Human Neuroscience 9:141.
Lenggenhager, B., T. Tadi, T. Metzinger, and O. Blanke. 2007. Video ergo sum: Manipulating bodily self-consciousness. Science 317 (5841): 1096–1099.
Locke, J. (1698) 1975. An Essay Concerning Human Understanding. Ed. P. H. Nidditch. Oxford: Clarendon Press.
Longo, M., F. Schüür, M. P. M. Kammers, M. Tsakiris, and P. Haggard. 2008. What is embodiment? A psychometric approach. Cognition 107 (3): 978–998.
Marr, D., and H. K. Nishihara. 1978. Representation and recognition of the spatial organization of three-dimensional shapes. Proceedings of the Royal Society of London 200 (1140): 269–294.
Martin, M. 1995. Bodily awareness: A sense of ownership. In The Body and the Self, ed. J. L. Bermúdez, A. J. Marcel, and N. Eilan, 267–289. Cambridge, MA: MIT Press.
Merleau-Ponty, M. 1968. The Phenomenology of Perception. London: Routledge.
Moseley, G. L., N. Olthof, A. Venema, S. Don, M. Wijers, A. Gallace, and C. Spence. 2008. Psychologically induced cooling of a specific body part caused by the illusory ownership of an artificial counterpart. Proceedings of the National Academy of Sciences of the United States of America 105 (35): 13169–13173.
Murray, C. D. 2004. An interpretative phenomenological analysis of the embodiment of artificial limbs. Disability & Rehabilitation 26 (16): 963–973.
Noel, J.-P., C. Pfeiffer, O. Blanke, and A. Serino. 2015. Peripersonal space as the space of the bodily self. Cognition 144:49–57.
O’Shaughnessy, B. 1980. The Will: A Dual Aspect Theory. Cambridge: Cambridge University Press.
O’Shaughnessy, B. 1995. Proprioception and the body image. In The Body and the Self, ed. J. L. Bermúdez, A. J. Marcel, and N. Eilan, 175–203. Cambridge, MA: MIT Press.
Petkova, V. I., and H. H. Ehrsson. 2008. If I were you: Perceptual illusion of body swapping. PLoS One 3 (12): e3832.
Petkova, V. I., M. Björnsdotter, B. Gentile, T. Jonsson, T.-Q. Li, and H. H. Ehrsson. 2011. From part- to whole-body ownership in the multisensory brain. Current Biology 21 (13): 1118–1122. doi:10.1016/j.cub.2011.05.022.
Scepkowski, L. A., and A. Cronin-Golomb. 2003. The alien hand: Cases, categorizations, and anatomical correlates. Behavioral and Cognitive Neuroscience Reviews 2 (4): 261–277.
Serino, A., A. Alsmith, M. Costantini, A. Mandrigin, A. Tajadura-Jimenez, and C. Lopez. 2013. Bodily ownership and self-location: Components of bodily self-consciousness. Consciousness and Cognition 22 (4): 1239–1252.
Simpson, D. C. 1974. The choice of control system for the multimovement prosthesis: Extended physiological proprioception (EPP). In The Control of Upper-Extremity Prostheses and Orthoses, ed. P. Herberts, R. Kadefors, R. Magnusson, and I. Petyersen, 146–150. Springfield, IL: Charles C. Thomas.
Soechting, J. F., and M. Flanders. 1989a. Sensorimotor representations for pointing to targets in three-dimensional space. Journal of Neurophysiology 62 (2): 582–594.
Soechting, J. F., and M. Flanders. 1989b. Errors in pointing are due to approximations in sensorimotor transformations. Journal of Neurophysiology 62:595–608.
Tsakiris, M. 2010. My body in the brain: A neurocognitive model of body-ownership. Neuropsychologia 48 (3): 703–712.
Tsakiris, M., and P. Haggard. 2005. The rubber hand illusion revisited: Visuotactile integration and self-attribution. Journal of Experimental Psychology 31 (1): 80–91.
Tsakiris, M., G. Prabhu, and P. Haggard. 2006. Having a body versus moving your body: How agency structures body-ownership. Consciousness and Cognition 15 (2): 423–432.
Vallar, G., and R. Ronchi. 2009. Somatoparaphrenia: A bodily delusion. A review of the neuropsychological literature. Experimental Brain Research 192 (3): 533–551.
van Erp, J. B. F. 2008. Absolute localization of vibrotactile stimuli on the torso. Perception & Psychophysics 70 (6): 1016–1023.
