The bodily self refers to the phenomenal experience of being an experiencing subject (i.e., a phenomenal self) bound to a physical body, which gives rise to the dual nature of the body (Husserl 1950; Gallese & Cuccio this collection, p. 2). The unified experience of being a bodily self can be decomposed into different aspects, including the experience that we identify with a particular body (self-identification or body ownership), the experience that the self is situated in a specific spatial location (self-location), that we take a specific experiential perspective at the world (first-person perspective), and that we are the authors of our actions, including having control of attentional focus (agency; (Blanke 2012; Ehrsson 2012; Jeannerod 2003; Metzinger 2003).
In their paper, Gallese & Cuccio highlight the relevance of mirror mechanisms, in particular related to processing in the cortical motor system, to the sense of body ownership and the sense of agency, in particular in the context of action and action observation:
This minimal notion of the self, namely the bodily self as power-for-action […], tacitly presupposes ownership of an action-capable agentive entity; hence, it primarily rests upon the functionality of the motor system. (this collection, p. 10)
However, recent philosophy of mind and cognitive neuroscience research reveals the crucial role of spatial aspects of the bodily self, consisting of a first-person perspective and self-location. In this section I shall compare the brain network contributing to spatial aspects of the bodily self with the brain network underlying body ownership and ask: Do these neuroimaging results support the proposal that motor resonance is a primary mechanism underlying all aspects of the bodily self? What is the relationship between the neural networks of the bodily self and social cognition? Which functional associations can be derived from this?
The phenomenal experience of being a subject is associated with a spatial location, which typically is the space of the physical body (see also Alsmith & Longo 2014; Limanowski & Hecht 2011). However, there are exceptions to these prototypical states of the bodily self in neurological disorders and experimental illusions pointing to a specific set of brain regions involved in spatially linking the phenomenal self to the physical body.
Which brain mechanisms link the phenomenal self to the physical body to give rise to the dual nature of the body as lived body and as physical object? Research in neurological patients who have had out-of-body experiences (OBE) shows that damage or interference with the right TPJ can lead to dissociations between the bodily self and physical body (Blanke et al. 2004; Blanke et al. 2002; De Ridder et al. 2007; Ionta et al. 2011). During an OBE, patients typically experience a disembodied self-location in elevation above their physical body, and an altered first-person perspective that originates from an elevated location in the room and is directed downwards at the physical body (Blanke et al. 2004; Metzinger 2009). These patients do not identify with their physical body but with an illusory double outside of the borders of the physical body. At the phenomenological level, self-location and the first-person perspective are often experienced as having their spatial origin in the same position. However, during OBE there are instances where self-location can be dissociated from the first-person perspective in different sensory modalities (De Ridder et al. 2007). Further evidence from asomatic OBEs and bodiless dreams suggests that a phenomenal first-person perspective may be reducible to a single point in space (Windt 2010). In fact, vestibular hallucinations systematically preceded OBEs in patients with sleep paralysis, i.e., a motor paralysis characterised by the transient inability to execute bodily actions when waking up from sleep (Cheyne & Girard 2009), showing further dissociations of the spatial location of the bodily self and the physical body and links to sensory processing. These studies seem to suggest that the first-person perspective and self-location may depend on different neural mechanisms (Blanke 2012).
OBE in epileptic patients can be induced by subcortical electrical stimulation of a specific intensity at the TPJ. However, stimulating the same brain region with either lower or higher stimulation intensity induces bodily sensations (including vestibular, visual, somatosensory, kinesthetic sensations) without inducing an OBE (Blanke et al. 2002). These observations gave rise to the idea that the spatial aspects of the bodily self are based on the accurate integration of multisensory signals (i.e., which was perturbed by electrical stimulation in the patient in Blanke et al. 2002, which are sensory signals from personal space to sensory signals from the external environment Blanke et al. 2004).
These clinical observations in patients were corroborated by different full-body illusion experiments in healthy subjects, such as the so-called “body-swap illusion” (Petkova & Ehrsson 2008; Petkova et al. 2011; van der Hoort et al. 2011), the “full-body illusion” (Ionta et al. 2011; Lenggenhager et al. 2009; Lenggenhager 2007; Pfeiffer et al. 2013; Pfeiffer, Schmutz & Blanke 2014), and the “out-of-body illusion” (Ehrsson 2007; Guterstam & Ehrsson 2012). In these experiments, healthy subjects receive conflicting signals about the spatial location of their body and of the temporal synchrony of exteroceptive and interoceptive signals, including somatosensory, cardiac, and vestibular signals that at the same time are applied to a virtual or fake body seen by the subject (Aspell et al. 2013; Ionta et al. 2011; Pfeiffer et al. 2013; Pfeiffer et al. 2014). For example, in the full-body illusion, synchronous stroking of a virtual or fake body seen from a distance can induce the feeling in participants that they are more closely located to the position of the virtual or fake body, and that they experience and increase of ownership for the seen body. The brain regions involved in these spatial experimental manipulations of the experienced bodily self most consistently involve the right TPJ region, but also draw on somatosensory and visual regions that process the sensory inputs (Blanke 2012; Ionta et al. 2011; figure 1b in black). Recently, several studies have manipulated visual and vestibular signals about the direction of gravity, affecting self-location and perspective and thus showing that those visual spatial cues affect our subjective experience of the first-person perspective (Ionta et al. 2011; Pfeiffer et al. 2013). These authors presented images on virtual-reality goggles showing visual gravitational cues, similar to the visual perspective during an OBE showing a scene from an elevated spatial location and a visual viewpoint directed downwards into the room. At the same time the somatosensory and the vestibular signals received by the participant, who was lying on the back, suggested that the physical body was oriented upwards with respect to veridical gravity. Thus the visual gravity cues (i.e., downwards) and the vestibular gravity cues (i.e., upwards) were in directional conflict. When the full-body illusion was induced under these conflicting conditions, participants reported subjective changes in their experienced direction of the first-person perspective (upward or downward) in line with experimentally-induced multisensory conflict (Ionta et al. 2011; Pfeiffer et al. 2013).
A different brain network encodes experimental manipulations of another aspect of the bodily self: body ownership. This was shown by the body-swap illusion (Petkova & Ehrsson 2008; Petkova et al. 2011), during which the participant views from a first-person visual viewpoint the body of a mannequin or another person. Thus no conflict between the visual spatial coordinates of the participant’s physical body and the visually-perceived location of the mannequin is presented. However, conflicting sensory information about the shape, gender, size, or overall spatial context surrounding the virtual body were presented that typically prevented feeling ownership of the virtual body. If under these conditions visuo-tactile stroking on the abdomen of the participant and the virtual body was synchronously administered, an illusion of ownership for the body emerged, reflected in increased responses to threatening the mannequin. In different variants of the body-swap illusion subjects reported experiencing and adopting different sizes of both the virtual body and the contextual environment (Petkova & Ehrsson 2008; Petkova et al. 2011; van der Hoort et al. 2011). Neuroimaging experiments of the body-swap illusion show activation of the vPM and IPS regions, notably without involving actions made by subjects or performed by the virtual body (Petkova et al. 2011). These brain regions are key nodes of the mirror mechanism network of ES (see Serino et al. 2013). For a recent review see figure 1b.
Although the neuroimaging evidence so far suggests that distinct brain regions encode the spatial aspects of the bodily self and body ownership (Blanke 2012; Serino et al. 2013), the ensemble of those bodily self-encoding regions closely matches the brain regions relevant for social cognition (compare in figure 1a with figure 1b). These empirical data indeed suggest that the bodily self and social cognition are encoded by at least overlapping neural circuits supporting the proposal of ES that neural capacities to control and monitor the own body are used in understanding others.
These neuroimaging data suggest particular functional associations between different aspects of social cognition and the bodily self. In particular, the brain network of ES anatomically overlaps with regions encoding experimentally-induced changes in body ownership during the body-swap illusion (figure 1a‒b in gray), which involves spatial congruence of the observational viewpoint and position of the fake body and the participant’s body. A second association can be observed between the brain network of AS and the brain regions encoding spatial aspects of the bodily self, as manipulated during the full-body illusion (figure 1a‒b in black). During the latter, the position and observational viewpoints of the virtual body and the participant’s body are in spatial conflict, and thus closely resemble social interaction settings.
Based on these functional and neuroanatomical observations, I propose that ES seems to contribute to the bodily self and social cognition in a way primarily related to the sense of body ownership and agency. However, ES does not account for multisensory spatial representations that relate the physical body to the bodily self in space. These spatial aspects of the bodily self are encoded by brain regions outside of the brain network of ES, and rather resemble those brain regions relevant for coding the spatial configuration of attention (or awareness, according to AS).
Because two crucial aspects of the bodily self, i.e., self-location and the first-person perspective, are encoded in the TPJ region, and full-body illusions show that they can be manipulated without action or motor manipulations, it seems implausible that ES as based on motor resonance is the primary brain mechanism underlying the bodily self. Instead, the brain networks coding self-location and the first-person perspective, which overlap with brain regions proposed to encode spatial aspects of an attention schema (see figure 1), seem to contribute to at least an equal degree to both the bodily self and social cognition. Thus, ES seems to be a necessary but insufficiently “primary” brain mechanism underlying the bodily self and social cognition.
I do not mean to imply that these are independent processes, because it is possible that they cooperatively work together (Graziano & Kastner 2011). However, I think that Gallese and Cuccio’s claim of a primacy of motor resonance underlying the multifaceted aspects of the bodily self and social cognition is questionable on empirical and theoretical grounds.