Using our body feels so natural: all you have to do is have the intention to make a movement for it to happen. This ability, however, hides many complex processes that still occupy many neuroscientists, psychologists and philosophers in search of explanations of what causes the feeling of having a body.
This sensation, called “feeling of incarnation”, is described by Kilteni et al . as composed of three dimensions:
- agency, that is to say the feeling of being the author of body movements;
- bodily property, that is, the feeling that the body is the source of the sensations experienced;
- self-localization, that is to say the feeling of being located inside the body.
If these three senses seem to prevent the dissociation of our body and our “mind”, it is nevertheless possible to create the illusion of having another body. Indeed, as strange as it may seem, carrying out this experience is now a breeze thanks to Virtual Reality (VR). Most of us know this technology for its ability to transport us to a different place , but it also allows us to embody a different body .
This illusion is made possible by the multiple sensory stimuli provided by VR headsets and which modify our perception of the world. Immersed in a 3D environment, the user takes the point of view of an avatar who responds to his actions as if it were his own body, thus producing the impression that he belongs to him.
The possibility of embodying another body interests many researchers who see themselves opening doors to otherwise impossible experiments. The objective of my thesis is to understand how we perceive avatars to make their use more natural. One subject that particularly fascinates me is that of the influence of self-image on our perception of the world: are we changing the way we see things by changing our representation of ourselves? If this question seems to be philosophical, it turns out to be of growing importance for researchers in Human-Machine Interactions.
At the start of my thesis, I started by learning about the work in virtual reality that has lingered on this subject before me. Some have obtained very astonishing results associated with a phenomenon called “Protein Effect”: virtually changing a person’s skin color would lead to a reduction in ethnic biases in the medium term .
Other studies go even further and encourage behavior change for therapeutic purposes ( eating disorders , pain treatment , etc.). Even more surprisingly, researchers have succeeded in showing that it is possible to temporarily improve our ability to solve problems by having participants play as Albert Einstein . Unbelievable ! And why not use Michelangelo as an avatar to boost our painting skills, or Jimi Hendrix for better guitar improvisations?
Inspired by all these results, I decided to embark on the study of the feeling of incarnation. In particular, I wanted to explore how to set up such a feeling without having to immerse myself in a virtual world. Previous studies have shown that it is indeed possible to evoke this type of sensation towards a mannequin or a prosthesis . However, the possibilities for experimentation with real objects are limited and difficult to set up.
Sense of bodily ownership
This is why I became interested in Augmented Reality (AR): indeed, this technology makes it possible to see and interact with dynamic holograms integrated into our real environment, and in particular with animated 3D avatars. Little is known about the perception of avatars in this context. If this turns out to be similar to that in VR, then that would mean that the behavioral changes observed in a virtual environment could be reproduced and exploited directly within our daily environment.
A study by Škola and Lliarokapis seems to support this hypothesis. In their article, the authors compare the feeling of incarnation in different contexts by reproducing the famous illusion of the rubber hand . This illusion consists in giving the impression to the participant that the rubber hand placed in front of him is part of his body.
To create this illusion, an experimenter strokes the rubber hand exactly at the same time as the participant’s real hand, hidden under a sheet. If the participant reacts physically to a threat, for example by withdrawing his real hand following a fall of a knife on the dummy hand, then this confirms that he has strongly appropriated the hand.
In the study by Škola and Lliarokapis, the experience of this rubber hand is compared to that of virtual hands viewed in augmented reality and virtual reality. Their results do not seem to show a significant difference between the overall perception of the three conditions. However, the authors find a stronger sense of bodily ownership in the case of the rubber hand than in the case of the virtual hand in augmented reality, but not in virtual reality.
Although no noticeable difference between the conditions was found, this last result particularly intrigued me. Could it be that the mixture of real and virtual elements in augmented reality is causing this subtle variation in the sense of ownership? This would explain why no difference in feeling was observed between the rubber hand and the virtual hand in virtual reality since in these two conditions, the visual is homogeneous. If this turns out to be true, then the environmental context would be an influencing factor for the feeling of embodiment that has never been identified.
Ethical and medical issues
My first experience was to study this question. Using an AR device, I compared the feeling of embodiment of virtual hands in the face of progressive amounts of virtual objects embedded in the real world. Each session, participants saw in front of their virtual hands 1) virtual objects, 2) real objects or 3) the two types mixed.
The measurements obtained through the questionnaires indicate a slight variation in the feeling of incarnation. Hands in condition 3 (mixed objects) seem to have elicited a stronger sense of bodily ownership than in condition 2 (real objects). Correlations were also observed on the one hand between the appropriation of the hands of the avatar and the immersion of the user, and on the other hand, between the appropriation and the perception of the coherence of the virtual content. These results suggest that the perceived coherence of virtual content is subjective and plays a role in the feeling of embodying another body.
However, how to explain the difference in bodily property between conditions 2 and 3, and above all, the lack of difference between the other pairs of conditions? It is not yet possible to answer with certainty and further studies will be necessary to quantify this bias. Further research on this subject seems crucial to ascertain whether the “Protein Effect” can take place under such conditions.
The stakes linked to the reproduction of this phenomenon are considerable, in particular in the medical field: augmented reality being more easily accepted than virtual reality in its daily use, the “Proteus Effect” would make possible the integration of virtual prostheses. in the daily life of amputees to relieve their phantom pain. It could also be used in post-stroke rehabilitation of patients, or even in the development of treatment strategies for mental disorders such as anorexia.
Many other examples in the field of education, interactive cinema, stage art, video games or even sport could benefit from the incarnation of avatars in AR. More generally, a user could integrate the illusions of incarnation into his daily life to accomplish tasks more efficiently by choosing the appearance of his avatar according to them.
But as AR graphics render more and more realistic, many ethical questions are emerging: how acceptable is it to change an individual’s behavior and perception?
While the virtual incarnation can bring many benefits, it is the responsibility of researchers, content creators and distributors of AR systems to develop a code of conduct to prevent the inevitable abuses and psychological and social implications of the process. virtual incarnation. Thus, in the rest of my thesis, my mission will not only be to broaden our understanding of this fascinating phenomenon, but also to discuss the possibilities to frame it.