ScienceAlert reports scientists at Binghamton University have built an electroencephalogram (EEG) cap that displays 500 images to the user at a rate of two per second. By analyzing the brain’s response to being shown the set of images, the system, known as Brainprint, is capable of working out whether the user is authorized or not.
The images are designed to extract strong emotional responses from the user and include pictures of food and landscapes, intriguing words such as “conundrum” and famous actors. By learning its users’ normal response to being shown the pictures, Brainprint is able to detect when the person wearing the EEG cap isn’t an authorized user, denying access to the device.
Lead researcher Sarah Lazlo said: “When you take hundreds of these images, where every person is going to feel differently about each individual one, then you can be really accurate in identifying which person it was who looked at them just by their brain activity.”
A side effect of the process is that the regular user can also be locked out if they have something else on their mind while trying to login. As annoying as this could be, the team claims the problem is actually an advantage. If an attacker forced the user to unlock their phone while at gunpoint, they simply wouldn’t be able to until their excited brainwaves returned to normal with the gun lowered and the fear lessened.
Project engineer Zhanpeng Jin said: “The key idea is that we want to identify and recognise the individual person based on their inside thinking. Inside-brain activity is not visible to anyone else. Even more exciting is that we want to use a non-volitional response. That means even the user cannot be aware of it.”
When testing Brainprint with 30 different people, the program identified the current user with 100 percent accuracy, proving it could work as a future security method. There are still some major issues though and we certainly won’t be unlocking smartphones while wearing an EEG cap for some time to come.
Despite its unhackable nature, carrying an EEG cap isn’t likely to become a popular method of signing into devices. The design will need to be refined, requiring an overhaul of the technology used to measure brainwaves. The system is also slow, requiring the user to look at 500 images before they are granted access. The technology clearly still has a long way to go but represents another interesting look into what the “password” of the future may be.
The long strings of characters, phrases and words are currently in decline as new authentication techniques with extra security and convenience increase in popularity. The proliferation of fingerprint scanners across smartphones is now extending into workplaces, a quick way to login that can only be thwarted if an attacker forces a finger onto the sensor.
Camera-based methods, such as iris and retina scanners, are also seeing more widespread use and are beginning to appear on some smartphones. A recent report found that passwords will likely remain the most popular authentication technique until at least 2025 but the growing numbers of next-generation methods are likely to become more popular over the next few years.
