Virtual worlds are fun to explore, but what about virtual bodies?

It turns out we can quickly learn to control an avatar in the form of an animal if our movements are mapped on to its digital representation.

Lights in the outer ring could only be hit with the tail, which could extend about half a metre beyond what the avatar's hands could reach.

The volunteers then played a virtual-reality game in which they had to use the avatar's hands or tail to hit colored lights that lit up in rings around it.

None of the volunteers was told how to control the tail, but half of them were given the ability to move it by twisting their hips while the other half had a tail that moved at random.

William Steptoe and colleagues at University College London gave 32 volunteers an avatar with a tail and allowed some of them to control it by moving their own bodies.

Those with a controllable tail had their hip position tracked by cameras 60 times a second, with the readings mapped on to the position of the tail.

The team found that people who could control the tail learned quickly.

Steptoe thinks this because mastering a new body part requires the brain to give less priority to existing ones. Thank you.

But while they were learning, they temporarily became less competent with their hands.

By the end of the 10-minute game they were as good at hitting the lights with their tail as they were with their hands.