Research at SUNY Seeks to Know if Brainwaves Can Verify Identity

With the responsibility of educating nearly half a million students each year, it’s no surprise to anyone that SUNY has a relationship with brains. But many people aren’t aware of the full extent of this relationship. Researchers at SUNY don’t just want to shape minds, they want to understand how they work.

Professors at Stony Brook invented MRI, which is used to search the brain for blood clots. Research from SUNY Downstate Medical Center in 2007 suggested the first new model of brain function since the 1940s. SUNY College of Optometry works in diagnosing and treating head trauma, and research done in part by the University at Albany recently released a study that worked on “Brain-to-Text” messaging. It is this, and research like it, that inspired the creation of the SUNY Brain Network, one of the SUNY system’s six Networks of Excellence. SUNY Brain is an interdisciplinary, intercampus research group dedicated to furthering the exploration of neuroscience.

So with all this ground-work research happening in understanding the brain, it is no surprise that researchers at Binghamton University have made another huge stride in understanding and utilizing the brain’s unique structure.

The goal of the study was to assess the “uniqueness, collectability, and permanence” of a particular brainwave signal.

The Brainprint Study

This year a team at Binghamton released a study about a new technique called "Brainprinting" in the journal Neurocomputing. The goal of the study was to assess the "uniqueness, collectability, and permanence" of brainwaves. This means that the team was trying to discern whether brainwaves could be used as an identifiable biometric, like a fingerprint or retinal scan.

In order to uniquely identify a person, brainwave data must be stable over time in the same way that a fingerprint is unchanging. To determine if brainwaves could be used in this way, the Binghamton team designed an experiment that collected, measured, and analyzed participants’ brainwave activity. Volunteers who came to the lab to be a part of the study were asked to silently read a series of acronyms to themselves several times across six months. As they read, researchers recorded their brainwave reactions via electroencephalogram (EEG), a test that detects electrical activity in the brain using small, flat metal discs attached to the scalp. In particular they were monitoring semantic memory.

Semantic memory is memory for concepts and relationships between them. According to Sarah Laszlo, assistant professor of psychology and linguistics at Binghamton University, semantic memory is, for example, knowledge of what a "bee" is (the concept of bee) and knowledge that bees sting and make honey, (i.e., relationships between bees and what they do). "Semantic memory is different from ‘episodic memory,’" Laszlo told us, "which I think is what people usually think about when they hear the word ‘memory.’ An episodic memory is a memory for some experience or event that happened, like a memory of your last birthday."

As Laszlo explained, monitoring your brain’s reaction to semantic memory is beneficial because this type of memory is stable over time. The brain is plastic and changes over a lifespan, so researchers needed to choose a potential biometric that would be stable over long periods of time. Semantic memory is stable. This is because while individual semantic memories are fluid, the full semantic system is very consistent. For example, if I had just learned the meaning of "SUNY," this would be a new semantic memory. But it is only one new acronym, and the average vocabulary of a college educated person is approximately 50,000 words. As Laszlo clarified, "It's like adding a drop of water to a bucket: sure, the amount of water in the bucket is definitely different, but in an essentially imperceptible way. We try to get a sense of the shape of the whole network in the testing, and the network as a whole is less sensitive to new information than any single concept’s representation is. It is simply a drop in the bucket."

‌‌‌‌

The Result of Our Tests

After months of testing, the team at Binghamton came up with some very promising results. Over the course of the experiment, the accuracy of the “brainprint” actually improved. The research team theorized that this was because test subjects became more familiar and comfortable with the process, so the apparatus gave better data after some time. By the end, the test was able to identify each participant with 93 percent accuracy; a number remains surprisingly stable even over a period of six months. Although this is not quite the high accuracy by which a biometric can be well established and widely adopted, researchers are confident these numbers can be improved upon.

As for the application of this research, Zhanpeng Jin, an assistant professor at Binghamton University, said, “We tend to see the applications of this system as being more along the lines of high-security physical locations, like the Pentagon or Air Force Labs, where there aren’t that many users that are authorized to enter.”

This technology is perfect for high security because, as Sarah Laszlo explained, “If someone’s fingerprint is stolen, that person can’t just grow a new finger to replace the compromised fingerprint—the fingerprint for that person is compromised forever. Fingerprints are ‘non-cancellable.’ Brainprints, on the other hand, are potentially cancellable. So, in the unlikely event that attackers were actually able to steal a brainprint from an authorized user, the authorized user could then ‘reset’ their brainprint.”

With this exciting new research Binghamton is helping SUNY build a better, safer world.

Published June 2015