Mind reading, a concept previously relegated to futuristic fantasy novels and late-night Miss Cleo advertisements, may no longer merely fall in the domain of science fiction. Researchers at Carnegie Mellon University (CMU) recently reported their discovery of a method to view and subsequently analyze the precise brain processes that occur as a person thinks about a specific word. Scientists say they are getting closer and closer to pinpointing the basic images and ideas that flash through a person's mind at any given moment.
The project was headed by Marcel Just, professor of psychology and director of the center for cognitive brain imaging, and Tom Mitchell, professor of artificial intelligence and machine learning. The researchers scanned subjects' brains by using functional Magnetic Resonance Imaging (fMRI) technology, which measures changes in activity in various parts of the brain by tracing blood oxygenation levels.
"The oxygenated blood goes to the part of the brain where it's needed because of high levels of activity occurring there," Tufts Professor of Psychology Holly Taylor said of fMRI scanning. "If you compare the brain image during one action to the brain image during a second action, you can almost subtract one image from the other to try to see the differences in brain activity."
By using this method, the researchers said they were able to successfully identify which word, out of several possible options, a subject was considering. Just coined the process "thought identification."
Similar research is currently in the works in several other locations, including Japan.
Several Tufts faculty members, however, expressed doubts as to whether the CMU thought identification technology is widely applicable or even accurate.
"It's one thing to follow the brain activity when a person's thinking about a hammer, but it's another thing entirely to try to read brain activity to see if a person is considering criminal activity," Patrick Forber, assistant professor of philosophy said. "It's much more complex when you're working with abstract concepts."
Like Forber, Taylor is skeptical of the findings for several reasons — one being the potential ambiguity inherent in the analysis of fMRIs.
"It's how you choose to interpret what the blood oxygenation level differences mean that has the impact," she explained. "It's subjective. How you analyze the data can actually influence the results."
She added that she is unsure whether modern technology could even keep up with human thought.
"The fMRIs are good because they have a high spatial resolution, meaning you can get a detailed, fine-grain look at different areas in the brain," she said. "But they have a low temporal resolution. Thinking is a very rapid process,and fMRIs are slow, relative to how quickly thoughts occur."
According to Forber, there are two important perspectives regarding thought identification technology — the ethical and the philosophical.
"First, we have to look at whether or not what we're doing is ethical. Are our interventions cohering with whatever policies we think are ethically right?" he said. "But then you also have to consider the philosophy of the mind — is it even possible to individuate the concepts of thoughts? It's a conceptual controversy."
Currently, subjects must be connected to machines in order to have their brains scanned, a process which is unlikely to occur without knowledge and consent.
"With the technology we have right now, the ethical implications are pretty thin," Forber said. "It's not exactly feasible to force someone to undergo the fMRIs. The [fMRI] machines are huge and extremely unwieldy; it would be very hard for a person to go through that and not be aware of it."
But some research today focuses on advanced technology such as using the reflection of a light beamed through a person's forehead and into their frontal lobes as a lie detection device — in other words, obtaining peoples' private information without their awareness.
"It raises an interesting question. The right to privacy is not necessarily a standard part of the Bill of Rights. We're going to need to determine legal precedents," Forber said. "But right now, that issue is still distant for us."
In some countries, however, the question has already been posed and answered. According to CBS News, a woman in India was recently convicted of the poisoning of her ex-fiancé after a brain scan revealed that she was allegedly aware of several circumstances regarding his murder.
Forber, however, cautioned against placing too much faith in machines. "Even within psychology, the usefulness of fMRIs is debated," he said. "Some people are worried that the technology and the resolution abilities just aren't precise enough to see the true and deep parts of the brain. In 15 years, we could look back on all this as a passing fad."
Many Tufts students were both impressed and intimidated by thought identification technology and its potential uses.
"I'm so incredulous that scientists today have these abilities. It's amazing to think of what's possible now," freshman Lauren Godles said. "It's a fascinating technology, but at the same time, I can't really think of many applications where mind-reading would be moral."
Sophomore Kevan Mamdouhi agreed that the developing technology is impressive. "It's incredible," he said. "We have come so far. We used to be able to detect lies on a basic level and now we can actually identify thoughts. It's amazing."
Although he was in favor of continuing to research thought identification methods, Mamdouhi, a biochemistry major, advocated for certain limits regarding the new technological initiatives.
"The ability to scan minds needs to be regulated by the government," he said. "You can't just have people running around using this as they see fit. But if it's under control, then this should absolutely be developed further. It's just undeniably cool."
