Body Swapping

Teh_Shadrin · Dec 6, 2008

Ghost in the Shell

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0003832

Found this on slashdot. Very interesting. Basically, some scientists have managed to convince experiment subjects that the body of a mannequin was their own, and even managed to continue the illusion when they shook hands with themselves.

In this experiment, the experimenter was wearing a specially designed helmet equipped with two CCTV cameras mounted in such a way that they presented the viewpoint of the experimenter (Figure 6). In turn, the participants stood directly opposite the experimenter, wearing the HMDs, which were connected to the CCTV cameras on the experimenter's head. Thus, the participants were facing the cameras. The participants were asked to stretch out their right arm and take hold of the experimenter's right hand, as if to shake it. This set-up allowed the participants to see their physical bodies from the shoulders to slightly above the knees. Hence, they could clearly recognize themselves and distinguish between their own arm and the arm of the experimenter. During the experiment, the participant and the experimenter were asked to repeatedly squeeze each other's hands for two minutes. In the illusion condition, the participant and the experimenter squeezed their hands in a synchronous manner, whereas in the control condition they squeezed each other's hands in an alternating rhythm, with the experimenter returning the squeeze in a semi-random manner. We tested the set-up in pilot experiments with ten participants (none of whom participated in the experiment reported here). Interviews conducted immediately after these initial experiments demonstrated that this set-up evoked a vivid illusion that the experimenter's arm was the participant's own arm and that the participants could sense their entire body just behind this arm. Most remarkably, the participants' sensations of the tactile and muscular stimulation elicited by the squeezing of the hands seemed to originate from the experimenter's hand, and not from their own clearly visible hand. In six other participants we also observed that this illusion worked well when the cameras were tilted downwards so that the participant could see the torso, legs and both arms of the experimenter's body during the manual interaction.

Also, gender seemed to make no difference. Both males and females reacted similarly to the male mannequin, and to the female experimenter.

The body-swap illusions worked well even though the mannequin or the other person looked different from the participant. In the first experiment there was no significant difference in rating scores between male and female subjects in the synchronous illusion condition, despite the fact that we only used a male mannequin (N = 32, p = .613, F(1,223) = .257, ANOVA) (Figure 8). Similarly, in the second experiment, male and female subjects alike were able to accept the arm of the female experimenter as their own. Further, we compared the threat-evoked skin conductance responses between males and females after threatening the new artificial body. To obtain sufficient numbers of males and females to enable a statistical comparison of the SCR, we pooled the data from the synchronous and asynchronous conditions where the stimulation was applied on the abdomen in experiments three and four. We found no significant difference in the illusion related SCR between males and females (p = .952, F = .004, Two Way Repeated Measures ANOVA). These observations suggest that gender identity, and differences in the precise shape of the bodies, are not important factors for perceiving a body as one's own

It also did not work with a non-human looking object.

We also conducted an experiment to examine our prediction that the body would need to look like a human to be experienced as one's own. Pilot experiments suggested that the illusion did not work with objects that do not resemble a human body, such as boxes, chairs and tables. Thus we conducted an experiment where we applied threats to the mannequin or to a rectangular object of the same size, after a period of synchronous or asynchronous visuo-tactile stimulation to the respondent and the object. This experimental manipulation was also important because it allowed us to eliminate the potentially confounding factor that associated learning in the synchronous condition could have caused the differences in SRC responses in the previous experiments. The results of the data analysis revealed that the threat-evoked SCRs were significantly stronger in the synchronous condition with the mannequin than in the synchronous one with the rectangular object. (N = 12, p = .008, two-tailed t-test) (Figure 5). For a third time we reproduced a difference in the SCRs between synchronous and asynchronous stimulation of the humanoid body (N = 12, p = .04, two-tailed t-test), without any such difference being observed in the conditions with the rectangular object (p = .819, two-tailed t-test). Thus, people can only experience human-like bodies as part of themselves. In addition the differences in the SCR between synchronous and asynchronous conditions are highly specific for the illusion.

Slashdot says "The effect could prove useful in virtual reality applications and in robot technology", but aside from practical applications, this is very interesting theoretically to me.

Log in or Sign up

Body Swapping

Teh_Shadrin New Member

Share This Page

Log in or Sign up

Body Swapping

Teh_Shadrin New Member

Share This Page

Useful Searches