Password

Could we ‘think’ the passwords of the future? Study finds brain‑reading tech is cheap, accurate and secure

The idea that we might ‘think’ passwords instead of typing them sounds like science fiction – but a team of UC Berkeley School of Information researchers has proved that it can work, using existing ‘mind reading’ headsets.

The idea that we might ‘think’ passwords instead of typing them sounds like science fiction – but a team of UC Berkeley School of Information researchers has proved that it can work, using existing ‘mind reading’ headsets.

The idea that we might ‘think’ passwords instead of typing them sounds like science fiction – but a team of UC Berkeley School of Information researchers has proved that it can work today, using existing ‘mind reading’ headsets.

The researchers used a Neurosky Mindset brainwave reader, an $100 EEG (electroencephalograph) device which scans brainwaves via a single contact on the forehead – and tested whether thoughts could work as passwords for a computer system. Subjects were scanned as they performed mental tasks such as singing a song of their choice (in their heads), or imagining moving a finger up and down, to generate a distinctive brainwave ‘password’.

The system worked with more than 99% accuracy – so much so it could used as a cheap, secure authentication system, the researchers claim.

Even in tasks where all the subjects did the same thing, such as moving a finger up and down, each person created sufficiently different signals for the computer to differentiate between users.

“We find that brainwave signals, even those collected using low-cost non-intrusive EEG sensors in everyday settings, can be used to authenticate users with high degrees of accuracy,” said the researchers.

“Other than the EEG sensor, the headset is indistinguishable from a conventional Bluetooth headset for use with mobile phones, music players, and other computing devices.”

The tricky part, say the researchers, could be finding mental tasks which people are comfortable with – when left to their own devices, users chose ‘private thoughts’ that were difficult to repeat. Choosing a song seemed popular with subjects, the researchers say, and worked effectively.

To Top

Pin It on Pinterest

Share This