The Beeblebrox Illusion: scientists convince people they have three arms

By Ed Yong | February 23, 2011 5:00 pm

“He relaxed and spread his two arms lazily across the seat back. He steered with an extra arm he’d recently fitted just beneath his right one to help improve his ski-boxing.” – Douglas Adams, The Hitchhiker’s Guide to the Galaxy

Zaphod Beeblebrox is just one of several characters from science-fiction and mythology to have extra arms. It’s a common enough trope, but would it actually work in real life? Could the human mind, which is so accustomed to controlling two arms, cope with a third or fourth one? According to Arvid Guterstam from the Karolinska Institute, the answer is yes. By placing a rubber right hand next to a person’s real one, and stroking both at the same time, Guterstam managed to convince people that they had a third arm.

Guterstam works in the lab of Henrik Ehrsson, who has spent many years studying how we come to feel that we own our body. It’s such a natural aspect of our lives that it seems bizarre to question it. But Ehrsson’s work shows how easy it is to mess with this sense of bodily ownership.

With a combination of cameras, virtual reality headsets and well-timed stroking, Ehrsson has managed to convince volunteers that they’re having an out-of-body experience, hovering above their real selves. He has persuaded amputees that a rubber hand was actually their own. And he has fooled people into thinking that they have swapped bodies, even with a person of the opposite gender or a mannequin. Giving people the feeling of owning three arms is just the latest of his sleights of hand.

Guterstam’s study goes against the common wisdom that our mental image of our bodies is constrained by our physical selves. For example, a part of the brain called the somatosensory cortex contains a literal “map” of our bodies, with different clusters of neurons matching up to different limbs and body parts. With a lifetime of two-armed experiences, it’s reasonable to think that our brain can’t cope with any more. Now we know that’s not the case.

The new trick riffs off the famous rubber hand illusion. Scientists can convince people that a rubber hand is actually theirs by stroking it at the same time as their real hand, which is hidden from view. But Guterstam found a very different effect when people could see their real hand.

He recruited 154 volunteers and asked each one to sit with their arms outstretched, and a realistic rubber arm next to their right one. A cloth obscured both arms above the wrist. Guterstam gently brushed the same finger of both right arms at the same time. Through questionnaires, he found that the volunteers “felt as if the rubber hand was [their] hand” and they “felt as if [they] had two right hands”. As previous studies have found, the illusion only worked if their real and fake fingers were brushed at the same time.

Guterstam didn’t just take their word for it – he also looked for physical giveaways. As we get emotionally aroused, we sweat and the extra moisture makes our skin better at carrying electric currents. After Guterstam stroked the real and rubber hands at the same time, he could increase the volunteers’ skin conductance by threatening the fake one with a knife. It was further evidence that they had genuinely adopted the third hand as their own. They still feel closer to their real hand (and are more perturbed if it’s targeted by the knife) but they accept the fake one too.

The illusion is the result of the brain’s attempts to resolve conflicting sets of information. It gets information from the skin that a hand is being stroked, but the messages from its eyes clearly tell of two stroked hands. The brain sorts out this apparent mismatch by accepting both hands. As a result, the volunteers think they have three.

Indeed, if the information from the senses doesn’t match, the illusion fails. Guterstam showed that volunteers aren’t fooled (or fooled less well) if the fake and real hands are stroked out of time, if the fake hand is turned the wrong way round, if a fake left hand is placed next to the right one, or if a foot is used instead of a hand.

These are timely results, especially as other scientists are busy developing increasingly advanced prosthetic limbs that can be controlled by thought. Most of these groups have focused on getting wearers to control the limbs accurately, but they also need to accept them as part of their bodies. Guterstam’s work suggests that this should be possible. Someone with a paralysed limb should be able to accept a prosthetic one, even if that raises their total number to three.

The prospect of getting extra mechanical arms might be enticing, but it’s still a long way off. Fooling people who are sitting at a table is one thing. Guterstam needs to see if he can keep up the Beeblebrox Illusion when people are actually performing tasks with the fake arm, and whether they can actually send different instructions to the real and fake limbs.

Reference: Guterstam, Petkova & Ehrsson. 2011. The Illusion of Owning a Third Arm. PLoS ONE

More on prosthetics: Monkey see, monkey control prosthetic arm with thoughts

More on illusions:


Comments (5)

  1. I quite like this research, but I’m not convinced talking about it as an illusion is the right way to go. There seem to be a couple of things going on;

    1. Those maps you mention aren’t static (or maps, per se, but that’s another story). They’re flexible and their current layout simply reflects what you’ve been up to recently; so when you pick up a tool that object gets incorporated almost immediately. So our cortical ‘sense of self’ is just ‘what we typically look like’ – it’s use-dependant, like the shape of muscles. The only reason the brain thinks we typically have two arms is that we do typically have two arms: change that, and the brain changes with it.

    2. This is related to the process of calibration: the various perceptual systems continuously calibrate and recalibrate each other to preserve accuracy. Left to its own devices, any perceptual system drifts and gets noisy. This process is ongoing and, with some limits, you can recalibrate the system quite extensively. That’s the point of the timing relationship between the visual and tactile information – these are informing each other, but only if they are in a suitable temporal range.

    The net result is that the only reason the brain thinks we have two hands is that a) we do but critically b) all our perceptual systems continuously inform the brain of this fact, actively maintaining that cortical organisation. Change (b) and cortex quickly (with some limits) reorganises.

    Anyway…I like all this recent work from these guys, I think it’s interesting, but I just get a bit grouchy about it being couched in terms of illusions. More interesting things are going on – what we ‘know’ about ourselves is actively maintained by the perceptual systems, and this isn’t even a little bit static.

  2. The illusion in the Hitchhiker’s picture at the top isn’t very convincing, unless the illusion involves the head of a distracted Paul McCartney…

  3. stonemason89

    Paul McCartney is alive, and he certainly doesn’t have two heads (or three arms)….

  4. But did he get any extra legs in the divorce?

  5. Georg

    A similar illusion (two noses) can be done without any
    Cross Your index and middle fingers (The middle goes over the index)
    and touch the tip of Your nose with the “valley” between those fingers.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Not Exactly Rocket Science

Dive into the awe-inspiring, beautiful and quirky world of science news with award-winning writer Ed Yong. No previous experience required.

See More

Collapse bottom bar