This Is Your Brain on Death
at Carnegie Mellon University, tries to take pictures of human thoughts. He was giving a talk about the way concepts are physically represented in the brain when his colleague David Brent, a psychiatrist, asked him if he could detect changes in the thoughts of people who are suicidal.
Just decided it was worth a shot. The two teamed up, and the answer, they concluded, was yes. In a paper published in late October in Nature Human Behaviour, their team investigated whether concepts like “death” and “life” are represented differently in the brains of people who had suicidal thoughts, compared with the brains of those who did not.
Just’s group looked at 34 young adults: 17 who were having suicidal thoughts and 17 who were not plagued by such thoughts. The researchers monitored each participant’s brain by functional magnetic resonance imaging (FMRI) as they were shown a series of words that represented different concepts. Some (“carefree,” “comfort,” “bliss”) were positive; others (“apathy,” “death,” “desperate”) could be related to suicide. Others (“boredom,” “criticism,” “cruelty”) were negative and unrelated to suicide.
The brain “signatures” of those with suicidal thoughts looked significantly different when shown the suicidal concept words. Of the words presented, “death” evoked the most dramatically different responses.
The researchers then turned a machine-learning algorithm loose on this data. The researchers “taught” the algorithm by showing it the FMRI data and the classification of the subject (whether the individual had thought about suicide) for all of the participants except one. After the algorithm was trained, the researchers gave it the FMRI data of the missing participant without telling it whether the person had been suicidal. The algorithm was able to determine whether that mystery
participant had experienced suicidal thoughts with 91 percent accuracy.
It’s important to note that this study doesn’t show a way of predicting who will and won’t die by suicide. But the research, Just says, is heading in that direction.
“It opens that possibility,” says Glenn Saxe, a psychiatrist at New York University’s School of Medicine who was not involved with the study. He notes that such a predictor “could really powerfully add to the tool kits psychiatrists have.” However, he cautions against overinterpreting the results of this paper, which do not describe a reliable means of predicting suicide.
For a matter as high stakes as determining whether someone will attempt suicide, accuracy is hugely important. Paul Sajda, a researcher in neuroimaging and computing at Columbia University, noted that
The algorithm was able to determine whether a participant had had suicidal thoughts with 91 percent accuracy.
while the algorithm is 91 percent accurate, the prospect of failing to recognize even one person as suicidal, or falsely labeling the individual as such, is disturbing. It also demands that researchers do a lot more work before thinking about using this as a diagnostic tool.
Sajda also wonders: Why use an FMRI when current psychological tests do about as good a job telling whether a person is suicidal? It was unclear to him what added value a test with an FMRI would provide.
Just and his team are aware of the potential and the risks here. He says many steps would have to be taken before this information could be safely used in assessing patients. For one, he’d like to replicate his findings with a larger number of people, although willing participants who are having suicidal thoughts are hard to come by. Also, sitting someone in an FMRI for 30 minutes is neither practical nor cheap.
He also stressed that anything that comes out of this research would be strictly an add-on to an assessment by a therapist or psychologist. “That’s the gold standard against which we compare it.”
As for any fears of a techno-dystopian scenario gone awry, Just insists that given how much work this diagnosis takes, “there’s no way this can be done against a person’s will…. You can’t point a little laser beam at them and find out what they’re thinking.”
THOUGHTS EXPERIMENT Just, top, and Brent found that words that could be associated with suicide triggered a different physical response in the brains of folks who’d had suicidal thoughts. SUICIDE ATTEMPTERS CONTROL GROUP