There is an old theory about monkeys, typewriters, and Shakepeare that suggests that, with an infinite amount of time, a money that randomly taps away at a typewriter could recreate the famous playwright's complete works. Now, scientists have actually done it - but not to study randomness. Instead, they are using it as a new way to translate our thoughts into written words. They use brain signals and convert them into movements across a keyboard.
Don't Miss: See the first leaked Black Friday 2016 Ad
The system, developed by researchers at Stanford University, amounts to typing with the mind. The team ran experiments on a group of monkeys and found that they could transcribe text at a rate of up to 12 WPM.
"Our results demonstrate that this interface may have great promise for use in people," said one of the researchers, Paul Nuyujukian. "It enables a typing rate sufficient for a meaningful conversation."
The monkeys didn't really learn how to read or understand, but they were trained to tap out letters based on what was on a screen. They completed Hamlet and moved on to passages from The New York Times.
It should have taken much longer, but Stanford researchers said that their new system is faster and more accuate than any that previous existed. The increase in WPM is attributed to the improvements in the algorithm converting thought to movement.
Brainwaves were read using a multi-electrode array which was then implanted into the monkey's brain. The system doesn't look for fully formed thoughts or even words, instead, it monitors the part of the brain that controls actions, like moving a mouse. Through the interception of these commands, the technology translates the thoughts into the movements of a mouse cursor on the screen so that letters can be individually picked out.
The next plan is to add auto-correct features.
"The interface we tested is exactly what a human would use," said Nuyujukian. "What we had never quantified before was the typing rate that could be achieved."
Don't Miss: Incredible Pokemon Gifts
The findings are published in the Proceedings of the IEEE.