Brain-computer interface technology extended to mobile phones

As the power of modern computers grows alongside our understanding of the brain, the development of brain-computer interface technology is quickly becoming a hotbed for research.  The brain-computer interface essentially creates a direct communication pathway between the brain and an external device.

A team of researchers at Dartmouth College led by Rajeev Raizada, research assistant professor at the Neukom Institute for Computational Science,  is currently working to extend the use of this technology into the field of mobile communication. Their findings are described in the senior thesis paper, “Neurophone: Brain-Mobile Phone Interface using a Wireless EEG Headset.”

Any brain-mobile phone interface needs to be able to detect neuronal signals from the cell phone user, interpret those signals, and then use them to drive mobile applications. The interface also requires a filter to distinguish between signals directed at the mobile phone and “background” noise or signals that do not pertain to the assigned task.

Based on these requirements, the Dartmouth researchers have developed a prototype application that automatically dials contacts in a cell phone’s address book based solely on neuronal cues from the user. These cues are sent via a non-invasive wireless EEG Headset from the user’s brain to the cell phone.

As the user scrolls through photos of the contacts in his address book, the EEG Headset detects and maps the electrical activity of his brain. When the user recognizes the photo of the person he wishes to call, he elicits a specific neuronal signal known as P300. This signal is distinguished from the others by the headset, processed, and then transmitted to the mobile phone where it activates the phone’s dialing mechanism.

The researchers conducted trials to assess the accuracy of their model in a variety of scenarios. Data was collected from subjects while they sat, stood, and walked. The dialing accuracy of the brain-mobile phone interface measured over 70 percent when the subject was sitting and about 60 percent when the subjects were standing. While these preliminary results are promising, there are still a number of practical challenges to overcome before brain-mobile phone interfaces can be widely implemented.

The technology is not yet cost efficient. In addition, the accuracy of the dialing mechanism decreases significantly when users are simultaneously conducting unrelated tasks like walking or listening to music. To correct this, the researchers are developing more advanced filters to discriminate against the neuronal “background noise” associated with these tasks.

Despite these challenges, the research team is confident that their brain-mobile phone interface is an “important development…that will open up new possibilities” not just limited to the field of mobile communication.