by Susana Martinez-Conde, PhD
Director, Laboratory of Visual Neuroscience
Barrow Neurological institute
Phoenix, Arizona

Some camera work, some stroking, and next thing you know you're out of your own body.
Photo courtesy Henrik Ehrrson

Sir Arthur Conan Doyle, best known as the creator of the coolly analytical detective Sherlock Holmes, was paradoxically a firm believer in the paranormal. His obsession with the supernatural fueled much of his (non-Sherlockian) fiction. In his short story "How It Happened," the protagonist wakes from a car crash. He is shocked by the experience but relieved to find his old friend Stanley standing beside him. The protagonist's view of the wreck is partially obstructed, and so he does not identify the inert body on the road. He then remembers that Stanley, whom he has not seen for years, died long ago.

"˜Stanley!' I cried, and the words seemed to choke my throat "“ "˜Stanley, you are dead.'

He looked at me with the same old gentle, wistful smile.

"˜So are you,' he answered.

Similar accounts of out-of-body experiences -- in which a conscious person sees his or her own body from a location outside the physical body -- have been reported in clinical conditions that disturb brain function, such as near-death experiences, epileptic seizures, drug abuse, stroke, and certain psychiatric and neurological disorders. Last year, two research groups induced out-of-body experiences in healthy participants with virtual reality techniques. The experiments, described last August in studies by H. Henri Ehrsson and Olaf Blanke and colleagues in Science, demonstrate that out-of-body experiences, previously confined to the realms of psychiatry, fiction and the occult, occur when the normal processing of sensory information is disrupted. This research provides an important tool to understand how the feeling of self is generated by the brain. Sherlock would approve.

Meet your virtual doppelganger

The experiments were conducted by research teams in the UK (H. Henrik Ehrsson) and Switzerland (Bigna Lenggenhager, Tej Tadi, Thomas Metzinger and Olaf Blanke). The participants wore virtual reality goggles connected to video cameras that filmed the participants' backs. Thus each participant saw his or her own body from the back.

But this trick alone did not induce an out-of-body experience. (And a good thing too. Otherwise you might have an out-of-body experience every time you check out your own backside in the fitting room at the mall). To complete the illusion, the scientists used two plastic rods to stroke synchronously, for 1 or 2 minutes at a time, the participant's back and the back of the virtual body. Next, the participants were asked to complete a questionnaire to evaluate their subjective perception of the illusion. Amazingly, they reported feeling as if they were being behind their physical bodies and looking at them from this location. The illusion failed when the stroking was asynchronous.

The results demonstrated that there are two key components to the feeling of being located inside the body. First, visual information from the first-person perspective provides indirect information about the location of one's body in space. The second factor is the detection of correlated tactile and visual events on the (illusory) body. Such multisensory correlations, together with the first-person visual perspective, determine the perceived location of one's whole body -- even if the correlated tactile and visual events are constrained to a small part of the body.

Don't walk towards the light!

Meanwhile, down in Switzerland, Lenggenhager and colleagues wondered if, following an out-of-body experience, participants might misjudge the location of their own bodies in space. To test this idea, they blindfolded the participants immediately after the stroking, then passively displaced them to a different position in the room. Then they asked them to walk back to their original location. Participants did not accurately return to their initial position, however. Instead they drifted significantly towards the previous position of the virtual body, suggesting that they had (at least partially) assigned the location of their selves to the virtual body. Such drift was not significant in the asynchronous stroking condition.

In a second experiment, the authors examined whether the illusion might depend on cognitive knowledge about bodies, and whether the drift towards the virtual body might be due to a general motor bias that happened to overshoot the initial position. To address these possibilities, they either presented the participants with their virtual own body (as in the previous experiment), a virtual fake body, or a virtual non-corporeal object (an elongated block) during synchronous or asynchronous stroking. Asynchronous conditions produced no illusion or drift. Synchronous stroking induced the subjective illusion for both the virtual own body and the virtual fake body, but not for the virtual object. That is, participants self-identified with both virtual bodies (their own and the fake), but not with the object. Moreover, the blind-folded participants showed significant drift towards both virtual bodies, but not towards the object. These combined results showed that the drift towards the virtual body was not due to a general motor bias but to the out-of-body illusion itself. Also, out-of-body experiences depend on the participants' knowledge about bodies: a non-corporeal object will not induce an out-of-body experience, whereas a bodily representation will, even if the body is not the participant's own.

I feel your pain

To provide further objective evidence for the illusion, Ehrsson "hurt" the virtual body by hitting it with a hammer and registered the electrical resistance of the skin of the (real) participants at the same time. The participants' skin conductance response (used by psychologists to measure emotional arousal) was significantly greater in the synchronous stroking condition (that is, in the presence of an out-of-body experience) that in the asynchronous condition (that is, in the absence of an out-of-body experience). Thus during an out-of-body experience, the participants responded emotionally to the threat of the hammer as if they were located behind their physical bodies.

A full-blown out-of-body experience?

Although the healthy participants reported seeing themselves from behind and misjudged the location of their bodies, they did not have the feelings of overt disembodiment that are typical of "full-blown" out-of-body experiences, such as those found in some patients with temporal-parietal damage. Lenggenhager and colleagues therefore proposed that other mechanisms in addition to the correlation of visual-tactile information (for instance, the correlation of visual-vestibular information) may be necessary to generate more complete transfer of the self to an illusory body. The authors speculate that the neural mechanisms underlying the spatial unity of self and body, as well as the disruption of such unity, may lie at the brain's temporal-parietal junction.

The experiments described here open a new research venue to discern the brain mechanisms generating our feeling of self. They also provide a scientific and rational explanation for supposedly paranormal experiences such as the out-of-body illusion, showing that this previously puzzling phenomenon can be replicated in the lab by simple experimental manipulations.

Susan Martinez-Conde is the director of the Barrow Neurological Institute's Laboratory of Visual Neuroscience, where she studies the neural code and dynamics of visual perception.