Lab Report: The Science of Tasting Beers

Feature by | Jul 2012 | Issue #66

Photo by Thomas H. Shellhammer

Like most coveted jobs, this gig isn’t quite the plum people think. For starters, there’s the glassware: flat-sided vials with clear and blue tops that look distressingly similar to urine samples. Then there’s the highly technical grading sheet, which shames even the geekiest beer journals. And it’s all done in a laboratory, in the middle of the afternoon.

“People think being on a tasting panel is like hanging out with your buddies drinking beer, and it’s not at all,” says Dr. Tom Shellhammer, professor of fermentation science at Oregon State University. “Yes, you get paid to drink beer, but it’s actually work.”

On a recent Friday afternoon, a tasting panel convenes in Shellhammer’s laboratory to taste Belgian and American sour beers. The names of the beers and the tasters must remain anonymous—this is serious science, bound for publication in a scientific journal—but we can reveal that the point of the study is to determine how the Old World and American Sours differ “using instrumental and sensory techniques,” Shellhammer says.

These trained tasters are, essentially, sophisticated testing machines made of meat. Culled from the town of Corvallis, Ore., home to the university, these OSU graduate students in food science, scientists and brewing or beer-related professionals were tapped more for their ability to pick out aromas flavors in beer. “They have to like beer enough to do the tests, but they’re not all beer drinkers,” says Shellhammer.

These trained tongues learn to pick out specific flavors, to the point that they agree on the difference between “rose” and “floral.” Most beer drinkers know “pine” and “green apple,” but these tasters all mark down “papaya,” “passion fruit” and “melon” and mean the exact same thing.

Tasting panels are trained by smelling the chemical aroma standards responsible for each flavor—as beer geeks know, banana flavor is isoamyl acetate and butter is diacetyl—in decreasing dilutions. They taste the isolated chemicals (aromatic compounds) added to polyethylene glycol, until they can detect small amounts of each.

Rather than learning the compound names, as brewers do, the panelists work together to develop a custom language for the group. “It’s really important to develop a vocabulary so people mean the same thing when they say it,” Shellhammer says. “When we do a tasting panel, we train them on that first, which is very time consuming but also very important. We have to really develop their palates.”

Then, it’s a matter of tasting. Again and again, every week, for months. Palates fatigue after about a half-hour, Shellhammer says, so tasters have to stop for the day. It’s a slow process but, gradually, patterns emerge from the data like the T-1000 rising from the vat of liquid metal to battle Arnold Schwarzenegger.

Nearly a year after a simple question was asked (“How different are Belgian and American sour beers, anyway?”) and this experiment planned, the results of this study will be submitted for peer review and to be published. The beers will remain anonymous, but there will be hard data showing how American brewers making Belgian styles, often with Belgian recipes and Belgian yeast, remain distinct.

This project—conducted by the OSU Fermentation Sciences Program, which trains students to wrangle microorganisms to make beer along with wine, cheese, yogurt, pickles and bread—is only one type of tasting panel. It’s the most thorough and formal, as it’s done for academic research, but most brewers put their product through similar tasting panel tests before releasing them.

Across the state, Full Sail Brewing in Hood River does rigorous lab and taste testing of every tank of beer before releasing it for bottling or barreling, says executive brewmaster Jamie Emmerson. “The art of science is in the recipe,” Emmerson says. “The science is making that recipe the same every time.”

There’s no shortage of equipment to help. The Full Sail lab looks like something in a hospital—not surprising, considering Emmerson came through the Siebel Institute back in 1987, the only person in his class not affiliated with a consistency-obsessed macrobrewer. Samples are pulled from the tanks and run through laboratory testing to make sure there’s no detectable drift from batch to batch. But the human olfactory and gustatory systems—our nose and mouth—are still far more sensitive instruments, also capable of seeing the forest for the trees in a way chemical testing can’t.

“It all comes down to taste in the end,” Emmerson says. “The lab is the final judge on all the technical aspects—is it clean, is the color right, is the bitterness is right—but all the things about the quality, the subjective piece, that has to come from taste panel.”

To do that, the brewery uses a total of 20 trained taste panelists from among the staff. All have been through testing, so they know what compounds they’re more or less sensitive to than the average person. Emmerson doesn’t look to stock the panels with so-called supertasters: “You realize that everyone is relatively equal as far as the palate goes, but there are just some things you’re better at and some things you’re worse at.”

Every tank release is tested by three tasters, and each panel is balanced with people who have different taste patterns that complement each other. “With everyone, there are a couple tastes you’re not sensitive to—you just don’t have the receptor to pick it up—and a few you’re hypersensitive to,” he says. “You wouldn’t want to have a taste panel of only people who weren’t as sensitive to bitterness, because then the beer could drift into being too bitter.”

They’re also looking to peg any potential off-flavors and eliminate the problems that caused them. “The idea of rotten eggs versus burnt match—those are two different sulfur compounds. So part of this is training people [on] what these things are, and why you would get these different smells—is it a process of fermentation, is it a process of cleaning—and you start to identify problems by off-flavors if you get them.

“Knock on wood, if we do our job right, there’s not a lot there that needs to be done, and the stuff that’s there is low level. You’re saying ‘OK, there’s something there that we need to work on,’ long before it gets to the point where it would be ‘Oh, shit,’” says Emmerson. “The corollary, which a lot of people don’t deal with because they’re too worried about the problems, is ‘what makes beer taste good,’ so if you’ve done your job right, you can worry about that.”