Interviewee: Read Montague,
Why does anyone pay money to poison their body, risk their health, and shorten their lifespan? Researchers at Baylor College of Medicine found that while smokers’ brains can recognize a losing gamble, that fails to affect their behavior.
“If you’re addicted to a substance then you pursue it in spite of all kinds of things that could happen in the future, in spite of negative things that could happen and in spite of positive things you might miss out on,” says study leader Read Montague.
“When you’re making decisions about how to behave and what to do and how to invest your time, things that might happen in the future can come to control your behavior, and can intervene on habits,” he explains. “So, for example, if you want to go to a job interview, maybe you decide to do your homework or you decide to put in a lot of extra work, and your need to eat or do something like that can be vetoed temporarily. What we found is that these fictional scenarios, scenarios about what might happen tomorrow, in normal people, people that aren’t addicted usually come to intervene on their behavior in this way. But in smokers, these fictive signals, which we can now measure neurally in the brain, are disconnected from their ability to make choices, and they’re driven completely by their habit-chasing systems.”
Montague and his team took MRI scans of chronic smokers and nonsmokers who played a fictional investing game.
“The market can either go up, in which case they’ve made money, or it can go down, in which case they’ve lost a little bit of money,” he says. “But after that, their choices are compared to choices that they might have gotten.”
To be able to make that comparison, they used mathematical methods from machine learning research to design and analyze the experiments.
“We used theoretical models from the computer science community from an area called reinforcement learning, which specifies how any computer program, any system, should learn from experience using feedback from the environment,” Montague says. “So without these models, we would never have been able to find these signals. We would never have been able to know where to look, or how to separate them from all the other things that we’re measuring in the brain.
By zeroing in on the brain regions that process these scenarios, the researchers could actually predict how nonsmokers would bet based on their consideration of the outcome of their previous choices.
“These might-have-gotten choices are what generate this fictional signal in their brain,” Montague says. “And so we separately traced what they actually gained, and what they fictively could have gained.”
But as published in the journal Nature Neuroscience these “fictive” signals had no influence on smokers’ choices.
“It’s the kind of system that keeps you searching and learning and trying for things. But for some reason in a chronic smoker’s brain it’s disconnected from controlling your behavior,” says Montague. He hopes these recently-developed methods might lead to new treatments.
“It’s possible that we could train you to use these, to listen to [these signals] more carefully, if we could give you feedback about how your brain is computing these things,” he explains. “And so what we hope to do is train people by giving them feedback, showing them in their brain what these signals are doing, teaching them to turn these signals up and listen to them more carefully when they make choices. We’ll see how that turns out in the future.”
The researchers are also planning experiments to figure out which comes first, the brain error, or the addiction. But they believe the findings in smokers will prove to be true for all addictions.
PUBLICATIONS: Nature Neuroscience, Advance Online Publication, March 2, 2008.
RESEARCH FUNDED BY: Kane Family Foundation, the US National Institute on Drug Abuse, the US National Institute of Neurological Disorders and Stroke, the Angel Williamson Imaging Center and the American Psychological Association.
Share on Facebook |
Tweet This |