Taste: Surprising Stories and Science About Why Food Tastes Good (28 page)

Roger and I used the gLMS method of scoring foods when we visited the Center for Taste and Smell in Florida. We were asked to set our upper limit—our highest score—by coming up with the “strongest imaginable sensation of any kind” that we had each experienced. They gave us a few examples. Many women rate childbirth as their strongest imaginable sensation. Those unlucky souls who have passed kidney stones often choose this as their strongest imaginable
sensation. Roger remembers the very moment during a dental procedure when his dentist hit a nerve. He asked the doctor for more anesthetic, but was told that he had already maxed out, so there was no way to up the dose safely. That five-minute hell was Roger’s strongest sensation. I had a harder time with mine and realized as a result just how cushy my life has been.

The next thing the gLMS administrators did was ask us to rate the brightness of the sun, and other things that we all experience. This gave them a couple of data points so they could “normalize,” or calibrate, our scores from this point forward.

Once we’d established the highest point on our scoring line, we were asked to start evaluating food. It is an odd thing to consider for the first time: how strong is the sensation of sweetness in a piece of chocolate compared with five minutes of anesthetic-free dental work? It takes some time to figure the scale out. We eventually got the hang of it as we rated everything from the five Basic Tastes to popcorn, lasagna, and grape jelly.

If you are going to evaluate more than two foods at a time, it’s very important to start off with the right one. Let’s use two foods as examples. First: salsa.

If you were conducting a salsa tasting, you’d want to sample salsas from mildest to hottest, because the heat from capsaicin (the compound that gives chile peppers their kick) tends to build up over multiple mouthfuls. With each mouthful, you also become more conditioned to the flavor of the salsa, so you want to be sure you’re making your judgment after the first bite or two. Malcolm Gladwell’s book
Blink
describes exactly what I do when I taste something in the lab. I try to make an initial
Blink
-style judgment about it: I go with my initial gut-based instinct in the first instant my senses are activated. I file that away as I continue tasting, and then see if anything changes as I keep eating. I need to take only one or two bites to make a comprehensive judgment about what I taste—thankfully, or I’d weight 300 pounds—but trust me, it has taken decades to develop this skill. Over the years I’ve come to trust my
Blink
assessment as the most accurate.

For our second example, let’s imagine that you are tasting four or five different cups of black coffee. Coffee has a famously bitter taste that also builds up over time, so it’s best to start with the least bitter of the samples, then move on to progressively more bitter ones. Save for last the thick, heavy brew that will put
hair on your chest. The same holds true for just about any category of food or beverage. The basic rule is to move from mildest to increasingly intense.

While the chefs and food technologists at Mattson know and practice this concept, they can’t prevent themselves from biasing me. When I am called into the food lab to taste a prototype, the first thing out of the developer’s mouth is usually, “Sample A has more . . .”—at which point I have to put my hand up and tell him to stop. Even though I’m acting as a human instrument, I’m unfortunately incapable of escaping human nature, which is to be biased by what you are told. So if Lin or Gorski tells me that Sample A is more salty than Sample B, it’s entirely likely that I will perceive this. I try very hard not to let this type of information influence me, but it’s really difficult, so I prefer not to hear it at all. I feel the same about reading wine tasting notes before trying a wine, reading a movie review before going to the theater to see it, and hearing a critic’s views on a new restaurant before I can form my own, unbiased opinion.

In the case of some really intense foods, such as hot sauce, we take another step when tasting. To ensure we can evaluate the full spectrum of tastes and flavors, we dilute the food with water. Consider the hot sauce sriracha (known to its diehard fans as “rooster sauce” because of the iconic rooster on the label of the best-known brand). Its label is designed to look Asian, though it’s made in America, and it’s ubiquitous at Asian restaurants. This intense, bright red sauce is thick, vinegary, and dominated by a deliciously fresh chile top note. It’s also scorchingly hot. A tiny dot on your tongue will put your taste buds out of commission for a few minutes. To avoid this, we dilute a small amount of it in water and then taste the sriracha-flavored water. By doing this we can swish the sample around to experience all of the volatile aromatics. And because it’s less hot, we can take a second or third taste. Professionals who taste alcoholic spirits use this same method. Because the alcohol level of most spirits is so high, it’s hard to distinguish subtle flavors, and it can be unpleasant to swish them around in your mouth—not to mention inebriating. By diluting spirits with water you can detect more than just the burn of the alcohol. True scotch aficionados will be likely to ask that you dilute their scotch with a bit of water, because they know how to get more flavor enjoyment from their drink of choice. This technique works even for foods you wouldn’t imagine diluting. Like chicken.

We at Mattson were working for a fast food restaurant chain whose research and development department had spent about six months developing their own twist on a very famous fried chicken sandwich. This famous fried chicken sandwich is, frankly, a work of art. It’s a boneless, skinless chicken breast, battered, then breaded with a simple blend of flour and spices. It’s fried to golden tenderness—not crispness—that results in a firm but juicy bite. Served on sweet, squishy white bread buns with dill pickles, the sandwich is a perfect balance of sweet, sour, salt, and umami. To me, it’s a crave-able, truly American indulgence right up there with apple pie, hot dogs, and bourbon-cornflake ice cream.

Our client’s goal was to develop something similar that they could offer in their restaurants for less money than the competition. They had been relatively successful in creating the sandwich, given the amount of time they’d put into the effort. Yet they were still about five percentage points away from parity preference, which is when consumers like both products the same. We use this as a measure of success when we’re trying to match, or reverse engineer, a food.

We frequently get requests from clients to “knock off Heinz ketchup” or “create an Oreo cookie clone.” Retail grocery stores, for example, might want to create a knockoff of a classic product so that they can sell it under their own store brand, usually at a lower cost. These projects are virtually impossible. As technologically advanced beings, we want to believe that there is a magical black box into which you can feed anything—a wine, a pharmaceutical, a voice, a tomato-based condiment—and it will spit out the complete instructions for how to re-create the product. Because you can do this type of matching with paint chips, people assume it must be possible with food.

In fact, reverse-engineering is always harder than you anticipate, because there are so many factors affecting the way a food tastes. Even if you know the fat, carbohydrate, and protein content of a food, the ingredient list, and its physical specifications (viscosity, pH, Brix, salt, and so on), that’s still not enough. The simple difference between tomato varieties can affect ketchup flavor significantly. At least for now, there are too many missing inputs to effectively create a perfect clone. The best method we have is trial and error. We whip up a batch of food, taste it as a team, then revise the samples and repeat the whole process—again and again.

At Mattson we won’t take on a project with a perfect match as the goal; it’s simply too difficult to accomplish in a reasonable amount of time for a reasonable amount of money. We
are
willing to take on a project with the goal of matching the degree of liking or achievement of parity preference. Consumers may recognize that Ketchup Sample 308 that we’ve created is not exactly the same as
Ketchup Sample 215, but they may like them both equally well. We are usually confident we can create an equally liked food, and we frequently do.

Back to the chicken. Our client was very, very close to achieving parity preference on the fried chicken sandwich they had developed to go up against the competition, but they weren’t quite there yet, even after rounds and rounds of trial-and-error development. Before they’d come to us, they’d also sent the product they were trying to copy for gas chromatography/mass spectrometry analysis. They’d analyzed the physical specifications of the product and hired a trained panel to conduct a descriptive analysis. But they could not close the gap between their sandwich and the competition’s original. So they asked us to help.

After much discussion about untried methods, I decided to approach this challenge the same way we evaluated the hot sauce. Others who had tasted the product at full strength had missed key flavor notes. As a consumer of the gold-standard product, I knew that the characteristic flavors are very subtle. Perhaps we could use the same dilution methodology to decipher what we were missing. Perhaps by diluting the ingredients in the sandwich, we could spread them out—like smearing paint across a canvas—to see what was in the trail. It was worth a try.

The first thing we did was taste the chicken sandwiches side by side. Then we tasted the fried chicken fillet without condiments or bun. Then we ran three additional taste tests using our chosen methodology.

First we put the battered and breaded fried chicken breast patty in a commercial blender with 200°F deionized water and whirred it to smithereens. Chicken smoothies, anyone? Then we filtered the smooth liquid puree through cheesecloth so that we were left with the very essence of the fried chicken breast fillet. What we had done was somewhat of a distillation, a steeping, or an extraction . . . of fried chicken.

Then we did the same for the chicken scraped clean of its breading. Scraping chicken breasts of their breading is the type of thankless task that interns at Mattson are asked to do. And just to make sure we’d covered all our bases, we whirred up a chicken breading smoothie extraction: just the scraped-off breading, none of the chicken. I know you’re salivating now.

This exercise turned out to be amazingly productive. After tasting fried chicken, naked chicken, and chicken breading in their natural states, we were almost sure we had identified the differences between the two sandwiches. But once we tasted the extracted broths, we were absolutely certain. Our client’s chicken breast was too high in umami and certain herbs, had too much of an oily mouthfeel,
and retained too much oil flavor from the frying process. These characteristics were obvious to us because of the method we’d used. And, no, my love for that chicken sandwich has not diminished even after I deconstructed it in this horrific way.

As a nightly wine drinker, I am a big advocate of drinking white wine on ice. The first reason is that the dilution principle holds true: even when something tastes fine on its own (single strength), you will savor more of it with a slight dilution. One or two cubes of ice result in the perfect amount of dilution to let you experience nuances you might otherwise miss in a crisp acidic, fruity wine that can be overpowered by bracing, heartburn-inducing acidity. The objective of adding ice to wine is not to lower the temperature of the wine; it’s to increase the dilution. This requires adding a just-about-right amount of ice. Too much ice will chill the wine to a point at which the volatiles are less active. The second benefit of subtly icing your wine is that you will dilute the amount of alcohol you’re consuming. This becomes more and more important as you get older. My fortysomething metabolism can’t handle the same amount of wine I enjoyed in my twenties and thirties. Of course, you can also add water to wine to achieve this effect, as the French have been doing for years.

Adding ice to very sweet white wines, such as dessert wine, ice wine,
¹⁴
and late-harvest wine, makes them more dilute, too, so that you can enjoy the fragrant floral and stone fruit flavors without the sometimes cloying sweetness. Ice wine makers in both the northern and southern hemispheres will cringe upon reading this next sentence, but it’s one of my favorite ways to drink sweet, high– residual sugar wines: I frappé them in a blender with ice and enjoy them with a spoon. Ice-blended ice wine. Yum.

This technique doesn’t work for all wines, because some white wines are better when they’re warmer. Cold suppresses and warmth liberates volatiles. Varietals such as chardonnay, which are low in fruit flavors, may release more of their nonfruit aromas (such as butter, oak, or vanilla) when they’re warmer
than refrigerator temperature. When you see wine drinkers cupping the bowl of a wineglass in their palms, they’re usually trying to increase the temperature to release more volatiles. The best way to figure out whether ice and/or dilution is a benefit or hindrance is to drop in a cube and taste.