Just put on a visor, move your arms, rotate your wrists and accept something quite absurd for a few minutes: two large feathered wings appear in place of your upper limbs. At first they seem like a game, a well-made video game fantasy. Then something more interesting happens. After a few days of training in virtual realitythe brain begins to respond to those wings differently, closer to how it responds to real arms. A study published in Cell Reportsin which 25 volunteers underwent a virtual flight program designed by researchers from Peking University and Beijing Normal University.
The experiment was born from an almost childish idea, in the best sense of the term: trying to understand what would happen if a person could fly without a plane, without an engine, without any real prosthesis hanging from the body. Yanchao Bi, a cognitive neuroscientist at Peking University, had spoken about it with Kunlin Wei, who directs the Motor Control Lab at the same university. From there the project took shape with the contribution of, among others, Yiyang Cai and Ziyi Xiong: to build virtual reality training in which human arms were replaced by wings and observe whether the brain was willing, at least a little, to take them seriously.
The body in front of the mirror
Participants wore VR headsets and motion sensors. In front of them, in a virtual mirror, they saw a body with large wings instead of arms. Every real movement of the upper limbs was translated in real time into the flapping of the wings: arms opening, wrists rotating, digital feathers responding. A small technological charade, of course, but built with rules precise enough to force the body to learn something.
The training lasted four sessions spread over seven days. Each meeting included a short familiarization phase and about 25 minutes of flight tasks: deflecting incoming balls, maintaining altitude, crossing rings suspended in the virtual environment. The software also included a simplified aerodynamic model. The downward beat generated lift, the upward beat produced drag. To ascend, participants had to extend their wings on the descent and contract them on the ascent. A kind of bird crash course, with less poetry and a lot more fMRI.
The behavioral results came quickly. In the tasks of navigating through the rings, the average score went up from 44.8% to 75.2%. The subjective feeling of control over the wings also increased. Some volunteers managed to “fly” almost immediately, others needed three or four sessions, but the improvement was visible throughout the entire process. The most curious part, however, was inside the head, literally.
The wings enter the map
Before and after the training, the researchers subjected the participants to brain scans while looking at different images: wings, human limbs, animal body parts, objects. The focus was on the occipitotemporal cortex, a region involved in the visual recognition of human bodies, limbs and silhouettes. After a week of virtual flight, this area responded more intensely to images of wings. Above all, in the right side of the brain, the neural patterns related to the wings were more similar to those associated with the upper limbs.
Translated without making it science fiction: the brain seemed to have moved its wings a little closer to the territory of the body. He treated them less as external objects and more as possible instruments of action. The important word here is “possible”. The study suggests that motor experience in virtual reality can change the way the brain interprets artificial body parts, even when those parts are not part of human evolutionary history. No one developed a new anatomy, no one left the laboratory with the instinct to glide from the balcony. Better to specify it, given the times.
The data becomes even more interesting because the shift was also seen in front of images of bird wings that were never checked during training. The participants, after using the virtual wings, also seemed to recognize other wings as potential “effectors”, that is, parts capable of acting on the world, a bit like hands, arms or legs. Not just feathered decorations, in short. Something he can do.
The scans also showed stronger communication between the right occipitotemporal cortex and some frontoparietal regions, areas involved in movement planning and integrating body signals. This increase in connection appeared when participants looked at wings, while it did not emerge in the same way when faced with other image categories. It is a technical detail, but it is significant: the virtual experience seems to have linked the vision of the wings to circuits more linked to action.
The brain remains cautious
The caution of the authors is fundamental. The wings did not become real limbs in the participants’ brain representation. After training, the neural patterns associated with wings still maintained similarities to those activated by tools or animal tails. In short, the brain did not throw open the door and say “please take a seat”. He closed the door ajar. And this alone is enough to make the study remarkable.
There brain plasticity it is precisely this margin of adaptation: the ability of the nervous system to reorganize itself based on experience, learning, the repeated use of a gesture or a tool. Here the experience was impossible in the physical world, yet coherent in the virtual world. The body saw wings where it expected arms, it moved them with a fairly stable relationship between gesture and consequence, it learned to use them to achieve a goal. In the long run, that consistency left a trace.
The theme goes beyond laboratory curiosity. Virtual reality can create bodies that human evolution never anticipated and allow the brain to inhabit them for a while. This can help understand how we integrate avatars, rubber hands, prosthetics, robotic limbs, artificial senses, and other body extensions. If the brain can treat two digital wings like something partially close to a limb, then it’s worth wondering how well it can adapt to devices designed to help people with amputations, sensory deficits, or new forms of human-machine interaction.
The study involves 25 people, short training, a controlled environment and a very specific task. It does not authorize flights of fancy on human beings ready to merge with any avatar. However, it shows one concrete thing: the image of the body, that silent map that makes us feel where we end and where the rest begins, is more manageable than it seems.
In the future we will probably spend more time in immersive environments, for work, play, rehabilitation, training, perhaps even for therapies that are yet to be well defined. At that point the question won’t just be how realistic virtual reality looks, how much a headset costs or how smooth the graphics are. The most uncomfortable question will be what the brain learns while we only think about playing. For now, four sessions, a few rings suspended in the air and two fake wings were enough. The rest was done by that stubborn old machine that we carry in our heads.
