Was it me who was speaking? Virtual body ownership can cause illusory agency
Normally we trivially distinguish between acts that we execute ourselves from those carried out by other people, our sense of agency over our own actions. Neuroscientists have developed theories of agency based on the intention to act followed by observation of the sensory consequences of the act - if these match then this is one element of the attribution of self agency to the act. Cause should precede effect, in other words the act and its consequences should follow normal common sense rules of causality. There should be no other obvious explanation for the effect, and a strong temporal binding between execution of the act and observation of the consequences of the act itself. We carried out an experiment showing that it is possible to trick the brain into the illusion that the participants were speaking when they were not. Participants wore a head-tracked head-mounted display and a full body motion capture suit that placed them in an immersive virtual reality. When they looked down towards their own body they saw a virtual body instead, which was also reflected back to them in a virtual mirror. In one experimental condition their virtual body moved synchronously with their real body movements, as captured by the motion capture suit. In another condition the virtual body moved independently of real body movements. From previous results we expected that those in the synchronous condition would experience the illusion that the virtual body was their body. This is referred to in the literature as the illusion of body ownership. After a few minutes of moving around with synchronous or asynchronous virtual body movements, the virtual body spoke some words. We found that those in the synchronous condition had the illusion that it was they who had spoken the words (even though they had said nothing). Moreover, the voice of their virtual body had a higher fundamental frequency than the real voices of the participants. Those who had experienced the synchronous moving body also later spoke with a higher fundamental frequency after their experience compared with before. Our conclusion is that a strong illusion of ownership over a virtual body can lead to the illusory attribution of agency to acts carried out solely by that virtual body. As we become represented more and more by online characters or operate tele-present robots, we must understand agency scientifically, to avoid the profound legal and ethical problems associated with self-attribution of agency to acts that w