Sarah Laszlo puts an EEG headset on a research participant’s head.
While your brain on drugs may be analogous to a fried egg, your brain on Instagram may be a super-secure form of identification, researchers report in a new study. Instead of fingerprints and retinal scans, the authors envision future authentication systems that probe a user’s brain while looking at random images, such as snapshots of Anne Hathaway or a slice of pizza.
According to the researchers, our brains create unique patterns of neural activity in the first few milliseconds as we process and respond to certain images, including images of food, celebrities and rarely used words. When the researchers collected and analyzed such patterns from 50 adult participants ranging in age from 18 to 43, they were able to create unique brain-based access codes that identified individuals with 100 percent accuracy and, as yet, seem impossible to duplicate. .
Study co-author Sarah Laszlo, a neuroscientist at Binghamton University in New York, told Ars:
We feel as humans that our brains are unique and our cognitive lives are unique, but that hadn’t been shown in the literature before… I think it’s just comforting and interesting to know, like really, your mind and your brain is different from everyone else’s. And this is measurable proof of that.
In addition, Laszlo and her colleagues argue that their quantifiable evidence of mental uniqueness could be used as a new biometric identifier that could one day replace other biometrics, such as retinal scans and fingerprints. Retinal scans once stolen cannot be changed to restore security. And fingerprints can be copied from photos – this recently happened to a German government official. Brain biometrics, on the other hand, can be easily modified by choosing different image patterns and are extremely difficult to surreptitiously capture and duplicate.
Others have previously tried to use brain activity for biometrics, Laszlo said. Previous efforts have focused on collecting brain waves from a resting mind or someone focusing on a user-generated key created by thinking about a specific memory or object. But, Laszlo explained, the data from those methods was too variable to accurately identify people. Being in a bad mood or distracted can disrupt brain activity patterns.
When Laszlo started the new study, she wasn’t convinced it would work either. But after some preliminary attempts returned with surprisingly good results, Laszlo dived into the research.
For the study, she and her colleagues placed participants one at a time in a quiet room and fitted them with headgear containing 30 brain sensors. The sensors would pick up on the continuous electrical chatter of their firing neurons, which appear as graphic squiggles in Laszlo’s measurements. Then the participants watched 500 images flash on a screen in front of them – each image remained for less than a second. (The researchers periodically asked participants to press a button during the process, just to make sure they were paying attention during the test.)
The photos were of things like food, celebrities, patterns, and unusual words. Laszlo said she selected the image categories primarily on intuition, focusing on the kinds of things that provoke personal reactions. People differ in what foods they like and hate, what they think of certain celebrities, and what weird words they know. But the short time frame to act on the images — on the scale of milliseconds — means the researchers are only capturing the participant’s automatic brain processing, something the participants can’t easily change at will.
With all the data, the researchers found they could sort background brain noise (such as bad mood and distraction) by the image-specific activity for each participant. And after analytical analysis, they mapped each participant’s biometric data back to brain responses for just 27 images, collected from just five sensors near the brain’s visual processing area. And that biometric could identify individuals with 100 percent accuracy in further tests, which continued for 10 months after the published study, Laszlo told Ars.
It’s possible that the brain’s biometrics may need to be reset or recalibrated after extended periods of time, Laszlo admits. But, she notes, we reset passwords every now and then anyway.
Laszlo also isn’t sure if the accuracy of the identification will decrease as she and researchers collect brain data from more people. Some people’s brain activity patterns may look similar. But, she argues, access to highly classified information or the most secure parts of the Pentagon — things that could warrant a brain biometric authentication — probably won’t have thousands or even hundreds of users, making the dilution of accuracy less of a concern. .
In new experiments, Laszlo and her colleagues try to mentally hack their system to prove it’s safe. So far, she tells Ars, it’s holding up.
The IEEE Transactions on Information Forensics and Security2016. DOI: 10.1109/TIFS.2016.2543524 (About DOIs).