Sensorimotor representation of the body multisensory integration and sense of self

The concept of self, something referred to by 'I', is central to psychology, but hard to study scientifically. The basis of self must be the body. The body is represented in the brain differently from external objects. First, I can directly move and control my body, but not other objects. Also, bodily sensory events feel different from perceptions of external objects. We ask how the bodily self relates to one's own actions and sensations, and how the brain's representation of the body modifies action and sensation processing? To do this, we need an experimental method that allows us to control whether an event is experienced as linked to my own body or not. We showed that 'Rubber Hand Illusions' (RHI) do precisely this. Subjects watch a rubber hand being touched while feeling a similar, synchronous touch on their own unseen hand. Perceptual regularity of the visual and tactile stimuli induces the illusion that the rubber hand is part of one's self: subjects incorporate the rubber hand into the mental representation of their own body. The effect is measured objectively as a drift in the perceived position of one's own hand towards the rubber hand. Results are compared to control conditions with identical visual and tactile stimuli are, but separated by a time delay between them. This asynchrony abolishes RHI. The project has 3 phases. We first compare sensory and motor aspects of sense of self using RHI. Sensory-RHI is induced by visual-tactile stimulation as above. Motor-RHI is induced by repeatedly making a simple finger movement which is repeated synchronously or asynchronously by the rubber hand, using an invisible motor. Sensory-RHI induces the feeling of 'ownership': that an external object is part of one's body. Motor-RHI induces a feeling of 'agency': that one can control the rubber hand directly by one's own command. By comparing the drift in the perceived position of one's hand towards the rubber hand, we can assess whether sensory or motor regularity makes a stronger contribution to the sense of bodily self. Is the self primarily sensory or motor? We next repeat the study in an fMRI brain scanner, and investigate which brain areas differ between synchronous and asynchronous conditions. This reveals brain regions responsible for sensory and motor aspects of self. We also look for areas of common overlap, which may integrate these aspects. The second and third phases of the project investigate which of the sensory and motor aspects of self is most fundamental. Is each aspect sufficiently powerful to influence the other? In phase 2, we assess whether the sensory self can influence motor systems. We induce sensory-RHI and then study whether a finger movement of the rubber hand (via the invisible motor) automatically activates the motor areas in the subject's brain, as if subjects were imitating or entrained by the movement that they see. A brain imaging study investigates the brain processes whereby ownership of the rubber hand can influence motor areas of the brain. In phase 3, we study the inverse relation, between motor-RHI and the sensory experience of the body. Motor-RHI is induced by voluntary movements, with synchronous or asynchronous movements of a video-projected hand. We measure the sense of touch in the hand with or without vision of a corresponding stimulus touching the video hand. Viewing touch should enhance tactile perception, and most strongly when the subject feels that the viewed hand is their own. We thus clarify whether sense of control also imparts a sense of ownership, and use fMRI to study the basis of this link in the brain. The project therefore provides an experimental and scientific way of studying aspects of the mind which are crucial to mental life, but have previously been inaccessible to science. The project links this key sense of self to specific brain processes, and to clearly-described experiences of sensation and action.

The neural representation of the body covers a range from low-level orderly projections of sensory and motor cortical neurons to the psychological representation of the self. This project attempts to understand the psychological and cognitive representations, associated with the body, by relating them to patterns of sensorimotor coding. Behavioural and neuroimaging studies manipulate the sensory, motor and visual experience of the body under precisely controlled conditions, and quantify the effects on subjective mental states and objective neural data. In particular, the experiments manipulate sensorimotor factors to identify the stimulus conditions associated with the feeling that the body belongs to 'me' (ownership) and that 'I' can control it (agency), and moreover estimate the neural bases of these representations. The research thus addresses fundamental questions both for the neural integration of multiple somatic signals, and for the subjective coherence of human bodily experience. An experimental approach must systematically manipulate the experience that subjects have of their own body. Since the body is 'always there' (William James), this has proved difficult. We will further extend our successful use of Rubber Hand Illusions (RHI) to manipulate body awareness. When subjects see a rubber hand being stimulated in synchrony with feeling similar stroking on their own hand, they experience a strong sense that the rubber hand is part of their own body. A correlate of the illusion is a perceived shift of the subject's hand towards the rubber hand (visual capture). We take this as quantitative proxy measure for the sense of bodily incorporation and body awareness. By comparing hand position shift due to either synchronised or asynchronous stimulation of the rubber hand, we showed that feedforward multisensory correlations play in important part in body awareness. Experiments 1 and 2 compare the intensity of the RHI in conditions where the correlated stimulation is either visuo-tactile, or visuo-motor. In the first case, the subject sees the rubber hand being touched while feeling touch. In the second, the subject moves their finger, and sees the rubber hand move likewise. Thus, sensory-RHI and motor-RHI correspond to ownership of and agency over one's own body respectively. These two aspects of body awareness are confounded in most previous studies. An fMRI study identifies the neural substrates of ownership and agency. The remaining experiments investigate whether ownership or agency is the more fundamental body representation. The general design involves inducing the sense that the rubber hand is incorporated into the subject's body by either purely sensory or motor means, and then assessing whether the altered body awareness transfers to influence motor or sensory processing respectively. Experiments 3-6 investigate whether pure ownership, as induced by sensory-RHI, is sufficient to modulate motor control. TMS measures of motor excitability (exp 3), behavioural (exp 4) responses and subjective judgements (exp 6) assess whether ownership suffices to induce agency. An fMRI study (exp 5) investigates how body ownership modulates motor circuits in the brain. Further studies test the inverse possibility, that the sense of agency over one's own body modifies purely sensory processing. In experiments 7 and 8, a motor version of the Rubber Hand Illusion (motor-RHI) is induced by subjects making simple finger movements while watching a video projection of their hand with or without an added delay. The sense of agency generated in the synchronous condition may modulate the integration of purely sensory visual and cutaneous sensations associated with that hand, demonstrating an effect of agency on ownership. The psychophysical and neural properties of this link are measured. The project thus gives a scientific, sensorimotor basis to neural body representation and the sense of self.