I've got a terrible sweet tooth. And I am kind of proud of it, in a way. Yeah, I CAN eat that whole chocolate cake. I'd even LIKE it. Honeycomb dipped in chocolate? YES PLEASE. There are very few sweet things that I'd refuse.

But should I really be ok with my sweet tooth? Could my sweet tooth correlate with something more sinister...a preference for alcohol?

Kareken et al. "A Preliminary Study of the Human Brain Response to Oral Sucrose and Its Association with Recent Drinking" Alcohol, Clinical and Experimental Research, 2013.


I can blame my sweet tooth on my parents, probably. Studies have shown that variability in preference for sweet things (though, to a greater or lesser extent, we all like sweet things), has a genetic basis.

But the sweet tooth doesn't go alone. In animals (mice especially), a preference for sugar in their water correlates with preference for alcohol as well. When you breed mice to make sure they drink alcohol (this is done to study alcoholism, for example), they also tend to really prefer sweet things, above and beyond mice that aren't so into martinis.

There is a correlation in humans, too. Humans who are more into sweet things are slightly more likely to abuse alcohol. But what is the basis?

The authors of this study wanted to look at the reward related systems of humans, and see how sweet taste might compare to alcohol drinking. They took 16 people, put them in an fMRI scanner, and then carefully sprayed their tongues with sugar water. fMRI looks at the blood oxygen levels in various areas of the brain. Higher blood oxygen levels are thought to correlate with increased "activity" of the brain (the idea being that more neurons in use means the area needs more oxygen). An example of this would be that your visual cortex will show increased blood oxygen levels when you are looking at something.

They then looked at the scans, looking particularly at areas of the brain associated with "reward", like the ventral striatum, and the orbitofrontal cortex. They compared the relative activation in these areas ("relative" is important here. In fMRI studies you compare the "activation" of one area to another area that is thought not to be in use, or to the area prior to the administration of the cue. In this case, they compared the blood oxygen activity rating in the area during the administration of water, and then after a spray of sucrose water). They showed that, unsurprisingly, people had higher blood oxygen levels in reward related areas when they tasted sucrose water.

But then they correlated the amount of blood oxygen levels in reward-related regions of the brain with the amount of alcohol that each subject DRANK.

Above you can see the graph correlating the blood oxygen signals in response to sucrose water with the amount of drinking the subjects did in the past week. They got a significant correlation between the fMRI response to sucrose water and the amount the subjects drank. The authors suggest that, because both sucrose water and alcohol achieve their rewarding results by increasing levels of dopamine in the brain, people who "need" higher levels of sweets might also prefer higher levels of alcohol. The idea is that maybe you could use a test like the sucrose one to tell who might need to stop drinking before things get out of hand.

I'm not quite so sure. It is definitely possible that those with a real sweet tooth (or a sweet brain, perhaps) might be more likely to drink more, and in the end, at higher risk for a drinking problem. The correlation is there.

But I don't think we can conclude that a sweet tooth means alcoholism from this study. They did find a correlation, but the sample size (only 16 people) was FAR too small. The variability there is high. With another 4 people, it might disappear entirely. And while a correlation is great...it doesn't show that these people who like sucrose would also respond more to alcohol. Would it have been terribly difficult to give them a squirt of alcohol in the mouth after the sucrose and look at the response there? I think that might have shown a great deal more.

And yes, you could eventually correlate the sweet tooth and the drinking...but what does that MEAN. After all, the liking for sweet things is probably the most evidence of just a higher sensitivity to rewarding stimuli, of which alcohol is another one. There are other things which correlate strongly with drinking behavior: impulsivity, high levels of anxiety, stress, genetic histories of alcoholism. Do these all also correlate with a love of sugar? The sucrose response is just one of many that may associate with alcohol response.

So while sucrose response and alcohol may correlate...well, there's more to alcoholism than correlation. There's more to reward from alcohol than a liking for sweets. And more work needs to be done before we have a sweet test for potential alcoholism. It will be interesting to see where future studies go.

Kareken et al. "A Preliminary Study of the Human Brain Response to Oral Sucrose and Its Association with Recent Drinking" Alcohol, Clinical and Experimental Research, 2013. DOI