Brain activity during abstinence may predict alcohol relapse risk

A small cluster of cells deep in the brain may hold a clue to one of medicine's most stubborn problems: predicting who will slip back into drinking after trying to quit. Researchers studying mice found that when animals were forced to stop drinking after months of voluntary access to alcohol, something shifted in their neurology. A region called the bed nucleus of the stria terminalis—a thumb-sized structure tucked beneath the brain's surface, known to researchers as the BNST—lit up with activity more than twice as bright as in mice that had never been forced to abstain. The mice with this heightened activity went on to drink alcohol even when it was laced with quinine, a bitter compound meant to discourage consumption. They drank more of it, too, suggesting that abstinence itself had rewired something fundamental in how their brains processed the desire to drink.

The finding emerges from a growing recognition among addiction researchers that abstinence, while medically beneficial, may paradoxically alter the brain in ways that increase relapse risk. The team gave mice long-term access to alcohol, then removed it entirely. When the animals were later placed back in the environment where they had previously drunk, they attempted to drink from empty spouts—a behavior tied directly to heightened BNST activity. But the most striking discovery came when researchers detected this brain activity before the mice even had a chance to attempt drinking again. The signal appeared in advance, like a warning light on a dashboard, suggesting that the BNST might serve as an early marker of who is most vulnerable to returning to alcohol use.

The stakes of this research extend far beyond laboratory mice. Alcohol misuse ranks among America's most consequential public health crises, yet it remains chronically underestimated in severity. In 2024, deaths linked to alcohol use were 4.5 times higher than deaths from opioids—a comparison that underscores how the nation's attention has been diverted elsewhere. Nearly 30 million Americans currently struggle with alcohol use disorder, a number that has effectively doubled since 1999. Over 80 percent of Americans age 12 and older drink at some point in their lives; roughly one in ten of those go on to develop a disorder. Yet clinicians have almost no reliable way to identify who will need help, and who among those seeking treatment will relapse.

Abstinence remains the cornerstone of most alcohol addiction treatment in the United States, even as harm-reduction approaches gain traction in opioid use disorder care. The FDA has approved medications for alcohol use disorder, but the sheer volume of people who need treatment—and the rising tide of diagnoses—has outpaced the field's capacity to help. A biomarker that could flag high-risk individuals before they relapse would represent a genuine shift in how treatment is approached. Instead of waiting for someone to fail, clinicians could identify vulnerability and intervene earlier, tailoring support to those most likely to struggle.

What remains unknown is substantial. Researchers do not yet understand the precise mechanism by which the BNST drives relapse behavior, nor what causes the activity to surge in the first place. The specific neurons within this region that encode the signal remain unidentified. The team is now using advanced neuroscience tools to manipulate BNST activity in mice, attempting to isolate its causal role in compulsive drinking. If they can establish that role—if they can show that quieting this region reduces relapse, or that amplifying it increases it—the path toward a human treatment becomes clearer. For now, the finding stands as a preliminary map of territory that addiction medicine has barely begun to explore: the brain's hidden machinery of relapse, waiting to be understood.