Machine-Learning Can Read Your EEG and Uncover Your Habits

Scanning your thoughts...

The future of identification authentication technology may, indeed, lie in brainwave scanning. It is a promising field, and its potential impact on daily life is intriguing, to say the least. Now, a pair of cybersecurity researchers from Texas Tech University claimed to have discovered another use for the technology.

Abdul Serwadda and Richard Matovu created a machine-learning system that compared two sets of EEG brainwave scans, one belonging to a group of identified alcoholics and the other from anonymous subjects. Using the machine, Serwadda and Matovu were able to correctly identify 25 percent of those people from the second group who identified themselves as alcoholics.

"We weren't surprised. We know the brain signal is so rich in information," says Serwadda. It's not 100%, but this early work is promising.

 

Identity confirmed

This technology could open up new possibilities in neuroscience. However, just as current identity-authentication technologies are susceptible to security breachers (fingerprints can be forged, facial recognition software can be fooled, etc.), brainwave scanning can also render you vulnerable.

Brainwaves do not come in organized, linear data patterns like fingerprints. Rather, they are a messy, vibrant puddle of personal information, all of which becomes accessible with EEG scans (though mapping out this information is still a difficult and imperfect process). Still, with the emergence of mainstream EEG scanning applications, such as portable EEG headsets used in gaming and other relaxation apps, retreating into the privacy of your thoughts may become difficult. 

“If you have these apps, you don’t know what the app is reading from your brain or what [the app’s creators are] going to use that information for, but you do know they’re going to have a lot of information,” Serwadda warns.

We may have to think about this one a bit longer.