Tapping into drivers' brain signals can cut braking distances and avoid car crashes, according to scientists.Researchers at the Berlin Institute for Technology attached electrodes to the scalps of volunteers inside a driving simulator.
The system detected the intention to brake, and cut more than 3m (10ft) off stopping distances, the team report in the Journal of Neural Engineering.
The team's next aim is to check the system in a series of road tests.
The 18 volunteers were asked to keep 20m (66ft) behind the simulated car in front, which braked sharply at random intervals.
Scientists used a technique called electroencephalograhy (EEG) to analyse the drivers' brain signals.
The system was able to pinpoint the intention to brake 13 hundredths of a second before the driver applied pressure to the brakes.
The team reported that at a speed of 100km/h (65m/h) the braking distance was reduced by 3.66 meters (12 feet).
Computer scientist Stefan Haufe told BBC News: "We know that any intention is generated in the brain. So it's no wonder that such things are visible in the brain.
"We were surprised it is so predictive. That is the thing!"
Lead investigator Benjamin Blankertz added: "It's quite easily explained by the fact that we can tap the driver's intention at the source of the build up of intention in the brain.
"It's a longer process, from the very first upcoming cognitive processes and intention building, until finally the muscles start the movement."
The volunteers also had the muscle tension in their lower legs analysed to detect the first signs of leg motion before they released the accelerator and pushed the brake pedal.
This data enabled the scientists to analyse the EEG information to determine which parts of the brain are key to braking. They improved the detection system accordingly.
'Point of no return' The Institute of Physics says this is the first time that EEG has been used to assist in braking.
The technique is, however, already used to help paralysed people control computers, prosthetic limbs and wheelchairs.
The researchers are planning to conduct road trials of their system to test its viability out of the lab.
But Benjamin Blankertz stressed that he suspects there may be some way to go before EEG can be used as a safety aid in real driving situations, not least because it requires the driver to wear a plastic cap with 64 electrodes covered in conductive gel.
This is uncomfortable, takes up to half an hour to fit, and the wearers have to wash the gel out of their hair afterwards. Smaller, more lightweight versions are in development.
The paper also mentions that wearers of EEG caps have to keep fairly still which is not always possible while driving, particularly when executing an emergency stop.
Dr Blankertz also said more work needs to be done on avoiding false alarms - to avoid the possibility that the machine could misread a drivers' brain signals and brake unnecessarily.
He said: "We need to investigate intention-building and decision-taking and self-initiated movement.
"Some recent research suggests that the outcome of free choices can be predicted from brain activity before the experimental subject is consciously aware of their intention.
"A technology that would make possible real time prediction of future decisions could be used to investigate how this relates to the so-called point of no return.
The team ultimately hopes to work with the automotive industry to combine their EEG technique with radar and laser systems that are used in some commercially available crash-avoidance systems, which detect obstacles such as walls, traffic signals and other vehicles.