You can lead a lab rat to sugar water, but you can’t make him drink—especially if there’s booze around.

New research published Thursday in Science may offer insights into why some humans who drink alcohol develop an addiction whereas most do not. After caffeine, alcohol is the most commonly consumed psychoactive substance in the world. For the majority of people the occasional happy hour beer or Bloody Mary brunch is where it stops. Yet we all know that others will drink compulsively, despite whatever consequence or darkness it brings.

The new research confirms earlier work showing this is true for rats; but it takes things a step further and supports a study design that could help scientists better understand addiction biology, and possibly develop more effective therapies for human addictive behaviors. Led by a team at Linköping University in Sweden, the researchers found that when given a choice between alcohol and a tastier, more biologically desirable sugar substitute, a subgroup of rats consistently preferred the alcohol. The authors further identified a specific brain region and molecular dysfunction most likely responsible for these addictive tendencies. They believe their findings and study design could be steps toward developing an effective pharmaceutical therapy for alcohol addiction, a kind of treatment that has eluded researchers for years.

A taste for sweetness is evolutionarily embedded in the mammalian brain; in the wild, sugar translates into fast calories and improved survival odds. For the new study, 32 rats were trained to sip a 20 percent alcohol solution for 10 weeks until it became habit. They were then presented with a daily choice between more alcohol or a solution of the noncaloric sweetener saccharine. (The artificial sweetener provides sugary-tasting enticement without the potential confounding variable of actual calories.) The majority of rats vastly preferred the faux sugar over the alcohol option.

But the fact that four rats—or 12.5 percent of the total—stuck with the alcohol was telling to senior author Markus Heilig, director of the Center for Social and Affective Neuroscience at Linköping, given the rate of alcohol misuse in humans is around 15 percent. So Heilig expanded the study. “There were four rats who went for alcohol despite the more natural reward of sweetness,” he says. “We built on that, and 600 animals later we found that a very stable proportion of the population chose alcohol.” What’s more, the “addicted” rats still chose alcohol even when it meant receiving an unpleasant foot shock.

To get a better sense of what was going on at a molecular level, Heilig and his colleagues analyzed which genes were expressed in the rodent subjects’ brains. The expression of one gene in particular—called GAT-3—was found to be greatly reduced in the brains of those who opted for alcohol rather than saccharine. GAT-3 codes for a protein that normally controls levels of a neurotransmitter called GABA, a common chemical in our brains and one known to be involved in alcohol dependence.

In collaboration with co-author and University of Texas at Austin research scientist Dayne Mayfield, Heilig’s team found that in brain samples from deceased humans who had suffered from alcohol addiction, GAT-3 levels were markedly lower in the amygdala—generally considered the brain’s emotional center. One might assume that any altered gene expression contributing to addictive behaviors would instead manifest in the brain’s reward circuitry—a network of centers involved in pleasurable responses to enticements like food, sex and gambling. Yet the decrease in GAT-3 expression in both rats and humans was by far strongest in the amygdala. “Figuring out the reward circuitry has been a fantastic success story, but it’s probably of limited relevance to clinical addiction,” Heilig says. “The rewarding effect of drugs happens in everybody. It’s a completely different story in the minority of people who continue to take drugs despite adverse consequences.” He believes altered activity in the amygdala makes perfect sense, given that addiction—in both rats and humans—often brings with it negative emotions and anxiety.

Much previous addiction research has relied on models in which rodents learn to self-administer addictive substances, but without other options that could compete with drug use. It was French neuroscientist Serge Ahmed who recognized this as a major limitation to understanding addition biology given that, in reality, only a minority of humans develops addiction to a particular substance. By offering an alternative reward (that is, sweet water), his team showed only a minority of rats develop a harmful preference for drug use—a finding that has now been confirmed with several other commonly abused drugs.

Building on Ahmed’s concept, Heilig added the element of choice to his research. “You can’t determine the true reward of an addictive drug in isolation; it’s dependent on what other options are available—in our case a sugar substitute.” He says most models that have been used to study addiction, and to seek ways to treat it, were probably too limited in their design. “The availability of choice,” he adds, “is going to be fundamental to studying addiction and developing effective treatments for it.”

Paul Kenny, chair of neuroscience at Icahn School of Medicine at Mount Sinai, agrees. “In order to develop novel therapeutics for alcoholism it is critical to understand not just the actions of alcohol in the brain, but how those actions may differ between individuals who are vulnerable or resilient to the addictive properties of the drug,” he says. “This Herculean effort to impressively map out a cellular mechanism that likely contributes to alcohol dependence susceptibility will likely provide important new leads in the search for more effective therapeutics.” Kenny was not involved in the new research.

Heilig and his team believe they have already identified a promising addiction treatment based on their latest work, and have teamed up with a pharmaceutical company in hopes of soon testing the compound in humans. The drug suppresses the release of GABA and thus could restore levels of the neurotransmitter to normal in people with a dangerous taste for alcohol.

With any luck, one of civilization’s oldest vices might soon loosen its grip.