I usually look for these beer and health stories as a means to keep telling myself that the stash is good for me, the stash is good for me. Now, however, it appears that while moderate consumption may sharpen the brain it could also be that the mopey college roommate of a pal was probably only acting just as nature intended:

Among the normal rats, the animals that consumed moderate amounts of alcohol fared better on both tests compared with the teetotalers. Rats on a heavy alcohol diet did not do well on object recognition (and, in fact, showed signs of neurotoxicity), but they performed better than their normal brethren on the emotional memory task. People often drink to 'drown sorrows,'" Kalev says. "Our results suggest that this could actually paradoxically promote traumatic memories and lead to further drinking, contributing to the development of alcoholism."

Oddly for anyone discussing science - but not, of course, odd at all where science and Victorian morality intersect - the summary of implications can actually include the statement "...the animals that consumed moderate amounts of alcohol fared better on both tests compared with the teetotalers..." from one scientist while also this statement from another:

"[This study] provides interesting evidence for a mechanism that may be operating at the NMDA receptor," he says, but quickly cautions: "It's better not to drink at all than to drink too much," as is also demonstrated by this study.

That logic would receive a flunking grade in most disciplines. Way to go Meir Stampfer, professor of nutrition and epidemiology at the Harvard School of Public Health. So even if some scientists have no idea what the science means, it is an interesting bit of research.