Scientists develop new “brain password” biometric using electrical response to images
Scientists at the University at Buffalo (UB) have developed a method of identifying an individual by reading their brainwaves as they view a series of pictures. The proposed biometric system, dubbed a “brain password,” is measured by a virtual reality headset, customized to read electrical activity through six electrodes.
VR headsets typically have 32 to 64 electrodes, but the scientists reconfigured one to record brain activity with three, while two more serve as grounds, and one as a reference point. The electrodes read the subject’s brain activity in three distinct areas with electroencephalography. The three areas are the intraparietal sulcus, which is associated with declarative memory, the inferior parietal lobule, which processes face recognition, and the temporo parietal junction, which controls reading comprehension, according to the announcement.
Subjects are shown images of an animal, a celebrity, and a phrase to activate the three areas of the brain, with all three shown in 1.2 seconds, and then repeated three more times, and found that the 4.8 second process identifies people with roughly 95 percent accuracy.
The benefit of this approach, according to the scientists, is that unlike other biometrics which can be spoofed if the template is compromised, it can easily be reset.
“To the best of our knowledge, this is the first in-depth research study on a truly cancelable brain biometric system. We refer to this as ‘hard cancellation,’ meaning the original brain password can be reset without divulging the user’s identity,” said Zhanpeng Jin, an associate professor at UB.
Researchers at Australia’s Edith Cowan University unveiled a cancelable system combining fingerprints and vein recognition earlier this year that they say prevents spoofing attacks.